Mass Spectrometry: A Foundation Course

Mass Spectrometry A Foundation Course

Mass Spectrometry A Foundation Course

K. Downard University of Sydney, Sydney, Australia

advancing the c hem ica I sciences

ISBN 0-85404-609-7

A catalogue record for this book is available from the British Library

0The Royal Society of Chemistry 2004 All rights reserved Apart from any.fair dealingfor the purpose of research or private study for non-commercial purposes, or criticism or review as permitted under the terms ofthe UK Copyright, Designs and Patents Act, 1988 and the Copyright and Related Rights Regulations 2003, this publication may not be reproduced, stored or transmitted, in any form or by any means, without the prior permission in writing of The Royal Society of Chemistry, or in the case of reprographic reproduction only in accordance with the terms of the licences issued by the Copyright Licensing Agnecy in the UK, or in accordance with the terms of the licences issued by the appropriate Reproduction Rights Organization outside the UK Enquiries concerning reproduction outside the terms stated here should be sent to The Royal Society of Chemistry at the address printed on this page. Published by The Royal Society of Chemistry Thomas Graham House, Science Park, Milton Road, Cambridge CB4 OWF, UK Registered Charity Number 207890 For further information see our web site at www.rsc.org Typeset by Refinecatch Ltd, Bungay, Suffolk, UK Printed by TJ International Ltd, Padstow, Cornwall, UK

TOCraig

“The occusionaljneness of line, the musterly distribution of musses” (from The Master (1895) by Israel Zangwill(lS64-1926))

Preface This book presents a broad coverage of the theory and application of mass spectrometry to provide the reader with an appreciation and understanding of the importance of mass spectrometry across a range of scientific disciplines. It is uniquely organised to enable a course or unit in mass spectrometry to be constructed at either the undergraduate or postgraduate level for students of a range of backgrounds and educational experiences where no single course can be deemed suitable for students of the physical, chemical, environmental, biological and medical sciences. It is published at a time when most available textbooks present an introduction to mass spectrometry, a broader treatise devoid of much detail or one that is focused on a particular area of the field. A large number of multiple author collections describing specialised disciplines, often inspired by a conference or workshop, together with new encyclopedic series have provided readers with up-to-date descriptions of mass spectrometry research and applications though usually in a less cohesive and accessible format. This has left a new scholar with some difficulty in comprehending the foundations, role and capabilities of mass spectrometry. This has motivated the construction of a new book on mass spectrometry that presents a broad treatise of the field across a wide range of scientific disciplines in a single accessible and affordable volume. Sufficient depth is presented throughout the book to enable students to understand the principles behind and the reasons for particular experiments, together with ample representations of mass spectral data and applications. Importantly, the book provides a reference text around which a series of university level courses can be constructed for the education of students with varied backgrounds, experiences and interests. The unique design of the book achieves this through the presentation of core sections that are common to all mass spectrometry experiments. These sections are coupled to content from other optional sections and specialised chapters dependent upon a student’s educational level, specialisation and interests. Recommended course structures are presented vii

...

Vlll

Preface

in the front of the book. At the same time, the organisation of the book is designed to present the field of mass spectrometry in a logical manner regardless of the course undertaken. Specialised chapters are included on organic mass spectrometry, ion chemistry, biological mass spectrometry featuring proteomics, mass spectrometry in medical research, the environmental and surface sciences and accelerator mass spectrometry. Large numbers of mathematical equations and derivations have been avoided and the theoretical description of mass spectrometry based experiments has been kept to a minimum. The absence of a large number of citations to the enormous body of published research on mass spectrometry was also deliberate, not so as to ignore the important work contributed by many scientists throughout the world, but rather to prevent the reader from being distracted by extensive annotations and references throughout the course of the text. Each chapter concludes with a list of key references and recommending reading material providing a springboard to further study. The author hopes that this book will assist with the teaching of mass spectrometry to the field’s future pioneers. Certainly, mass spectrometry education will remain an important exercise given the important, and in some cases essential, role that mass spectrometry plays in scientific discovery. Kevin M. Downard

Contents Guide to a Foundation Course in Mass Spectrometry Acknowledgements

XV

xvi

Chapter 1 Mass Spectrometry’s Beginnings

1

1.1 A Brief History 1.1.1 Early Pioneers and Cathode Rays 1.1.2 Positive Rays 1.1.3 The First Mass Spectra 1.2 Isotopes and their Implications for Mass Measurement 1.2.1 Discovery of Isotopes 1.2.2 Isotopes and Mass Measurement 1.3 Molecular Weight 1.3.1 Elemental Composition and Mass Accuracy 1.3.2 Nitrogen Rule 1.3.3 Double-Bond Equivalents Further Reading

1 1 1 3 4 4 4 6 6 7 8 9

Chapter 2 The Mass Spectrum

10

2.1 Concept of Charge and the Molecular Ion 2.2 Fragment Ions 2.2.1 Formation of Fragment Ions 2.2.2 Stability of Fragment Ions 2.2.3 Stabilising Effects 2.2.4 Quasi-Equilibrium Theory 2.2.5 Metastable Ions 2.3 Relative Ion Abundance 2.4 Mass Resolution 2.5 Mass Measurement and Accuracy Further Reading

10 11 11 12 13 13 16

ix

11

18 19 21

Con tents

X

Chapter 3

The Mass Spectrometer

3.1 Basic Components 3.2 Ionisation Techniques and Interfaces 3.2.1 Electron Ionisation 3.2.2 Chemical Ionisation 3.2.3 Coupling Gas Chromatography to Mass Spectrometry (GC-MS) 3.2.4 Field and Plasma Desorption Ionisation 3.2.5 Fast Atom or Ion Bombardment 3.2.6 Laser Desorption and MALDI 3.2.7 Spray Ionisation Methods; Thermospray 3.2.8 Electrospray Ionisation 3.2.9 Atmospheric Pressure Chemical Ionisation 3.2.10 Coupling Liquid Chromatography and Capillary Electrophoresis with Mass Spectrometry 3.2.11 Low Flow Rate Electrospray Ionisation - Nanospray 3.3 Mass Analysers 3.3.1 Time-of-Flight 3.3.2 Magnetic Sector 3.3.3 Quadrupoles 3.3.4 Quadrupole Ion Trap 3.3.5 Ion Cyclotron Resonance 3.3.6 Hybrid Instruments 3.4 Detectors 3.4.1 Faraday Cup 3.4.2 Electron Multipliers 3.4.3 Microchannel Plate Electron Multipliers 3.4.5 Array Detectors 3.5 Computer Acquisition of Data 3.5.1 Role of Computers in Mass Spectrometry 3S.2 Analog-to-Digital Converters 3.5.3 Data Processing and Interpretation Algorithms 3.6 Vacuum Pumps 3.6.1 Rotary Pumps 3.6.2 Diffusion Pumps 3.6.3 Turbomolecular Pumps 3.6.4 Cryopumps Further Reading

22 22 23 23 25 26 26 27 30 33 33 35 36 38 39 40 43 47 49 51 54 55 55 56 57 58 59 59 60 60 61 61 62 62 64 65

Contents

Chapter 4

XI

Tandem Mass Spectrometry

67

4.1 Basic Principles; Precursor and Fragment Ions 4.2 Dissociation Processes and Theory 4.2.1 Collisional Activation (CA) 4.2.2 Collisional Activation Theory 4.2.3 High (keV) and Low Energy (eV) Collisions 4.2.4 Charge Reversal and Stripping 4.2.5 Photon-Induced Dissociation (PID) 4.2.6 Surface-Induced Dissociation (SID) 4.2.7 Electron Capture Dissociation (ECD) 4.3 Tandem Magnetic Sector Mass Spectrometers 4.3.1 Mass-Analysed Ion Kinetic Energy Spectra (MIKES) 4.3.2 Linked Scans 4.4 Tandem Quadrupole Mass Spectrometers 4.5 Tandem Mass Spectrometry on Ion Traps 4.5.1 Tandem Mass Spectrometry on Quadrupole Ion Traps 4.5.2 Tandem Mass Spectrometry on FT-ICRs 4.6 Tandem Mass Spectrometry on TOF/TOF Instruments 4.7 Tandem Mass Spectrometry on Hybrid Instruments Further Reading

67 68 68 69 70 71 72 72 73 73 73 75 77 78

Chapter 5

84

Organic Mass Spectrometry

5.1 Accurate Mass Measurements 5.1.1 Calibrating the Mass Scale 5.1.2 Peak Matching 5.2 Fragmentation of Organic Molecules 5.2.1 Mass Spectral Databases 5.2.2 Location of Charge and Predictive Bond Fission 5.2.3 Homolytic Cleavage 5.2.4 Heterolytic Cleavage 5.2.5 o-Bond Cleavage 5.2.6 Rearrangements 5.3 Fragmentation of Organic Molecules by Compound Class 5.3.1 Hydrocarbons 5.3.2 Alcohols 5.3.3 Ethers 5.3.4 Amines 5.3.5 Aldehydes and Ketones 5.3.6 Carboxylic Acids, Esters and Amides 5.3.7 Halides

79 80 80 81 83

84 84 85 86 86 86 87 88 88 89 90 90 93 95 96 97 98 99

Contents

xii

5.4 Quantitative Analysis of Organic Compounds 5.4.1 Role and Choice of Quantitation Standards 5.4.2 Calibration of the Detector Response 5.4.3 Quantitative Analysis of Cotinine; Example of Selected Ion Monitoring Further Reading

100 100 101

Chapter 6

104

Ion Chemistry

103 103

6.1 Electron and Proton Affinities and Measurements of Gas Phase Acidity 6.1.1 Electron Affinity 6.1.2 Gas Phase Acidity and Proton Affinity 6.1.3 Gas Phase Acidity Measurements 6.1.4 Kinetic Method 6.2 Ion-Molecule Reactions 6.2.1 Types of Ion-Molecule Reactions 6.2.2 Rates of Ion-Molecule Reactions 6.2.3 Ion-Neutral Intermediate Complexes 6.3 Kinetic Isotope Effects Further Reading

104 104 105 107 108 108 108 110 110 111 112

Chapter 7

113

Biological Mass Spectrometry

7.1 Ionisation of Biomolecules and Biopolymers 7.2 Peptides and Proteins 7.2.1 Molecular Weight Analysis 7.2.2 Mass Mapping 7.2.3 Peptide and Protein Sequencing 7.2.4 Protein Structure and Folding 7.2.5 Protein Complexes and Assemblies 7.2.6 Proteomics 7.3 Oligonucleotides and Nucleic Acids 7.3.1 Identification of Modified Nucleosides 7.3.2 Sequencing of Oligonucleotides by Tandem Mass Spectrometry 7.4 Oligosaccharides and Glycoconjugates 7.4.1 Sequencing of Oligosaccharides by Tandem Mass Spectrometry 7.4.2 Exoglycosidase Digestion 7.4.3 Derivatisation Approaches: Oxidative and Reductive Cleavage to Identify Branching Further Reading

113 114 114 115 117 124 128 132 138 139 139 140 142 144 144 146

Contents

xiii

Chapter 8 Mass Spectrometry in Medical Research

148

8.1 Characterisation and Quantitation of Drugs and Metabolites 8.1.1 Introduction 8.1.2 Sample Preparation Techniques in Drug Discovery 8.1.3 Qualitative Analysis of Organic Drugs and their Metabolites 8.1.4 Quantitative Analysis of Drug Compounds and their Metabolites 8.2 Defining Metabolic Pathways with Mass Spectrometry 8.3 Characterisation of Drug Libraries by Mass Spectrometry 8.4 Drug Screening using Mass Spectrometry 8.5 Trace Element Analysis in Nutrition Further Reading

Chapter 9 Mass Spectrometry in the Environmental and Surface Sciences

148 148 149 149 153 154 155 156 157 159

160

9.1 Environmental Analysis 9.1.1 Heavy Metals and Elemental Analysis 9.1.2 Organic Pesticides 9.2 Isotope Ratio Mass Spectrometry 9.3 Portable Mass Spectrometers 9.4 Chemistry of the Earth’s Ionosphere 9.5 Mass Spectrometers in Space 9.5.1 Apollo Missions 9.5.2 Viking and Mars Express Missions 9.5.3 Composition of a Comet 9.6 Applications of Secondary Ion Mass Spectrometry to Materials Science 9.6.1 Depth Profiling 9.6.2 Analysis of Impurities 9.6.3 Reaction Catalysts Further Reading

160 160 162 163 165 167 168 169 169 170

Chapter 10 Accelerator Mass Spectrometry

175

10.1 Introduction 10.2 Ion Sources 10.3 Performance and Limitations of Radiocarbon Dating 10.4 Applications of Radiocarbon Dating in Archaeology and Cosmology

175 176 177

171 171 171 172 173

178

xiv

Contents

10.5 Biomedical Applications Further Reading

180 181

Appendix 1 Abbreviations used in Mass Spectrometry

182

Appendix 2

Isotope Masses and Abundances

185

Appendix 3

Comparison of Common Ionisation Techniques

197

Appendix 4

Comparison of the Performance of Mass Analysers

198

Appendix 5

Common Neutral Losses During the Fragmentation of Organic Compounds

199

Summary of Common Fragment Ions Detected for Organic Compounds by Class

200

Appendix 7

Gas Phase Acidity Data

20 1

Appendix 8

Amino Acid Residue Masses and Modifying Groups

202

Appendix 9

Mononucleotide Residue Masses

203

Appendix 6

Appendix 10 Monosaccaride Residue Masses

204

Appendix 11 Web Sites on Mass Spectrometry

205

Subject Index

206

Guide to a Foundation Course in Mass Spectrometry

Undergraduate Core

Chemistry

Physics

Biulugy

1.2.2 1.3 2.1 2.2.1 2.3-2.5 3.1 3.2.1-3.2.3 3.2.5 3.26 3.2.8 - 3.2.10 3.3 3.3.1 3.3.2 4.1 5.3

Ch 1 2.1 3.1 3.2.1 4.1 6.2 6.3 9.4-9.6 Ch 10

2.1 2.3-2.5 3.1 3.2.63.2.9 3.2.10 3.2.11 4.1 7.1 7.2.1-7.2.3

Medicine Envirunmental Sciences

2.1 2.3-2.5 3.1 3.2.1-3.2.3 3.2.6 3.2.8 3.2.9 Ch 8

2.1 2.2.1 2.3-2.5 3.1 3.2.1-3.2.3 3.2.5 9.1-9.3 9.6 10.1 10.3 10.4

All sections listed represent the entire section (with sub-sections). All subsections listed represent the entire subsection only. *optional sections and subsections should be added to the core material in the order that they appear in the text, not the order they appear in this table. ** postgraduate material should be taught in addition to, or as a supplement for, the undergraduate material dependent on the exposure of students to this subject matter at the undergraduate level.

xv

Acknowledgements I owe a particular gratitude to John Bowie for introducing me to the exciting field of mass spectrometry and for his support throughout my career. John’s internationally recognised research in gas phase ion chemistry instilled in me an early appreciation of the positive aspects of negative ion mass spectrometry beyond the analytical. I am grateful to many colleagues and students past and present, too numerous to mention here, in both the mass spectrometry community and further afield who have contributed to my own education and challenged my teaching of mass spectrometry. I also thank my wife and family for their love and support. Finally, it has been my pleasure to work with Janet Freshwater, Robert Eagling, Tim Fishlock and the entire editorial and publication team at the Royal Society of Chemistry. My thanks also go to Edward Abel, former president of the society, who catalysed this interaction. Their belief in, and support of, this project has made the book possible. Kevin M. Downard

xvi

CHAPTER 1

Mass Spectrometry’s Beginnings 1.1 A BRIEF HISTORY 1.1.1 Early Pioneers and Cathode Rays Mass spectrometry had its beginnings in experiments performed over a century ago. Scientists in the late 19th century began conducting experiments within evacuated glass tubes in order to gain some understanding of the nature of electricity. George Johnstone Stoney was the first to report that electricity has its basis in a particle, or an “atom of electricity” that he referred to as an electron. Stoney measured the charge of the electron in 1894 but it was left to Joseph John (J.J.) Thomson to measure the charge-to-mass ratio (elm) of the electron and estimate its mass at a thousand times less than that of a hydrogen atom. Thomson had developed an interest in the electron while investigating the passage of electricity through gases in his laboratory in Cambridge. Thomson believed that the stream of rays emitted from a negatively-charged cathode, known as cathode rays, consisted of these particles. He also proposed that the particles (which Thomson preferred to call corpuscles) were one of the bricks from which all atoms were built a controversial theory at the time. Thomson went on to describe his case in the book Corpuscular Theory of Matter published in 1907. -

1.1.2 Positive Rays

Some time earlier Eugen Goldstein, a scientist in Germany who had given the name to cathode rays and studied them for several decades, discovered that the presence of gases in cathode ray tubes also gave rise to rays that behaved very differently from cathode rays (Figure 1.1). Wilhelm Wein in 1898 was able to deflect these rays in the opposite direction to cathode rays using magnetic and electrical fields. He 1

Chupter 1

2 electric field plates

+-

plate

Figure 1.1 Representation of the cathode ray tube showing the deflection of positive rays (ions)

concluded that they were the positive equivalent of the negativelycharged cathode rays and carried “positive electricity”. Wein’s experiments, however, showed that these rays contained particles with masses much larger than an electron and of the order of a hydrogen atom. Fascinated by this positive form of electricity, Thomson improved upon the data of Wein by operating his cathode ray tube at lower pressures and showed also these positive rays could be deflected from a straight line by perpendicular electric fields onto a photographic plate (Figure 1.2).

Figure 1.2 Parabolic paths ofionisedforms of atoms and small molecules ivithin a cathode ray tube (Source: Fig. 5, plate 1, J.J. Thomson, Recollections and Reflections, G. Bell and Sons Ltd., London, 1936)

Mass Spec trome try ’s Beg innings

3

1.1.3 The First Mass Spectra

First with hydrogen, and later with other atoms and molecules of carbon, nitrogen and oxygen, Thomson discovered that each charged particle followed its own parabolic path based upon their detection on the photographic plate. He reasoned that no two particles would strike the plate at the same place unless they possessed the same velocity and chargeto-mass ratio. Thomson quickly realised that an inspection of the plate showed at a glance how many kinds of particles there were in the rays and that, by knowing the value of elm for one parabola, the values of elin for all the others could be deduced. In his recollections, published in 1936, four years before his death, Thomson recounted that the positive rays contained atoms and molecules of all gases, common elements or molecules present, and correctly suggested that the positively charged particles were formed by the loss of an electron. In other words, the positive rays consisted of charged atoms or molecules, known as ions. Thomson concluded that the positive ray spectra possessed many advantages over other approaches for chemical analysis. The number of components as well as their atomic or molecular weight could be measured from these spectra. He argued that by using long exposures the approach could be “exceedingly delicate” allowing for the presence of a trace of gas to be detected “too small in amount to be measured by any other spectroscopic method”. Another noted advantage was that the method was not dependent on the purity of gas analysed. Impurities merely appeared as additional parabolas in the spectrum and did not contribute errors to measurements of atomic or molecular weight. Thus the field of mass spectrometry was born and these important features, recognised by Thomson, remain to this day. Mass spectrometers are able to: (i) measure the atomic and molecular weights of charged species in complex sample mixtures, (ii) analyse compounds at extremely low sample levels, (iii) analyse compounds in a mixture without purification of that mixture. These advantages, and many others, considerably outweigh the few disadvantages described later such as the loss of the sample once it is analysed. But first the work of Thornson’s student Francis Aston is briefly reviewed. Aston’s work led to the discovery of isotopes which have important implications for mass analysis.

4

Chapter 1

1.2 ISOTOPES AND THEIR IMPLICATIONS FOR MASS MEASUREMENT 1.2.1 Discovery of Isotopes Commencing in 1909, Francis Aston accepted an invitation to work as an assistant to Thomson in Cambridge to study positive rays. It was during this period that improvements to Thornson’s cathode ray experiments were made and several new “mass spectrographs” were constructed. These instruments, forerunners of the modern mass spectrometer, enabled Aston to separate two isotopes of elemental neon, *‘Ne and 22Ne. The principle was extended to other chemical elements and led to the discovery of 212 naturally occurring isotopes. From this work, Aston formulated the so-called “Whole Number Rule” that states when the mass of the oxygen isotope is defined, all other isotopes have masses that are nearly whole numbers or integers. Carbon, rather than oxygen, is now considered the standard isotope mass upon which all other isotopes are measured.

1.2.2 Isotopes and Mass Measurement Isotopes are atoms of the same element with different numbers of neutrons in the atomic nucleus. These add mass but not charge to the atoms and molecules composed of them. Naturally occurring carbon, for example, is a mixture of two isotopes 12C and I3C that represent 98.9% and 1.1% of all the carbon on Earth. The individual carbon atoms have masses of either 12.00000 (assigned as a standard by the International Union of Pure and Applied Chemistry, IUPAC) or 13.003354mass units (u, formerly atomic mass units amu). The average mass of carbon is calculated as 12.011 u based on these masses and the natural occurrence of each isotope (98.9/100 x 12.000 + 1. M O O x 13.003 = 12.011). Isotope masses and relative abundances for all the common elements are provided in Appendix 2. If a mass measurement is made where the isotopes of each atom are separated, the mass will reflect a monoisotopic (or one isotope) mass. If the isotopes are unresolved, the mass will reflect an average mass. Note that the average mass value is larger than the monoisotopic mass for the lightest isotope since it contains a contribution from the heavier isotopes. Unlike Thornson’s early mass spectra, modern mass spectrometers record both the mass-to-charge ratios of ionised forms of atoms or molecules and also their relative abundance in the spectrum. Thus

Mass Spec t romet ry 's Beg innings

5

when the isotopes of an atom are separated or resolved in a mass spectrum, the relative intensities of the ions reflect their relative levels. To see this, consider the mass spectrum of an atom of chlorine. Chlorine has two isotopic forms 35Cland 37Cl.The mass spectrum for an ionised chlorine atom will therefore appear as shown in Figure 1.3 in a bar graph representation. 106

35

CI

106

CI,

70

72 %

50-

%

37

50-

74 0-,

I

1

40 m/z

060

I

I

70

80

m/z

Figure 1.3 Electron ionisation mass spectra of atomic and molecular chlorine

Note that the relative height or abundance of the 37Clisotope is one third of that of 35Cl.This is consistent with the relative abundances of the two isotopes which are 75.78% and 24.22% for 35Cland 37Clrespectively (Appendix 2). Consider a molecule of chlorine gas Cl,. A molecule can be formed either from two 35Clatoms, one 35Cland one 37Clatom, or two 37Clatoms. The mass spectrum of the ionised form of the molecule reflects this and shows three resolved ionic forms (at mlz 70, 72 and 74) with relative abundances of approximately 100, 64 and 10% respectively (Figure 1.2). These values can be derived theoretically by multiplying the relative abundance for each atom and normalising the values to 100. The relative abundance for chlorine molecules comprised of mixed isotopes must be multiplied by two in order to account for the two forms possible; that is 35C137C1 and 37C135Cl. It is a simple matter to calculate the relative height of isotope peaks for any particular molecule. In the case of an isotope of one integer unit of mass higher than the molecule (i.e. M + l), the relative height of this peak can be expressed as a percentage given by equation 1.1. In this equation, %I represents the percentage of naturally occurring isotope one mass

6

Chupter 1

unit greater than the common isotope for each element, E. The number of atoms of that element in the molecule is defined by n. ratio of (M + 1)/M peak = C, ( % I x nE) %

(1.1)

Hence for a molecule of ethanol (C,H,O), the ratio of the (M + l)/M isotope peaks will be equal to (1.08 x 2) + (0.015 x 5 ) + (0.038 x 1) or 2.27% based upon the percentages of naturally-occurring I3C, ,H and I7O(see Appendix 2). In general, the contribution of any element with two common isotopes (of a and b percent) to an isotope peak (M + N)/M is given by equation 1.2. ratio of (M + N)/M peak = N!(a)"-'(b)"/N(a)"

(1.2)

Hence the relative peak height of (M+2)/M for a molecule of Cl, is calculated to be 2(0.76)'(0.24)2/2(0.76)2= 0.076. Note that N! (or N factorial) equals the multiple N x (N - 1)x (N - 2) x . . . to x 1. 1.3 MOLECULAR WEIGHT

The molecular weight of a compound is the sum of the atomic masses for all atoms in the molecule weighted according to the relative abundances of their isotopes. For example, the molecular weight of a molecule of sucrose is the sum of the atomic masses for 12 or common sugar (C,2H2201,) carbon atoms, 22 hydrogen atoms and 11 oxygen atoms. The average molecular weight calculated from the atomic weights for these atoms is 342.299. If the isotope peaks for the compound were resolved or separated in a mass spectrum, the expected mass of the monoisotopic ions or ions containing only the lightest isotopes of each element would be 342.1 16. Measuring molecular weight is a fundamental activity of modern mass spectrometry, but as we shall see later, mass spectrometers are used for many other purposes including the complete structural characterisation of molecules, the quantitation of components within complex chemical and biological mixtures, fundamental investigations of ion behaviour and reactivity studies of molecular complexes, and even the radiodating of archaeological relics. 1.3.1 Elemental Composition and Mass Accuracy The manner in which accurate mass measurements are obtained is described later in Chapter 5 in the context of the instrumentation

Mass Spectrometry’s Beginnings

7

required, but assume for the moment that the molecular weight of a compound can be measured with a high degree of accuracy. Subject to the accuracy obtained, it may be possible to assign the elemental composition to a compound based solely on the molecular weight measurement. This is because the monoisotopic atomic masses for the elements are not exact integers; rather they have a fractional mass component. Consider for example a small organic molecule whose molecular weight based on a measure of the mass-to-charge ratio of its monoisotopic ions is 46.0419. This molecular weight value is consistent with a molecule of ethanol with an elemental formula of C2H,0, but not for instance a molecule of nitrous oxide NOz with a molecular weight of 45.9929. The mass accuracy required to distinguish these two molecules can be calculated as one-half of the difference between their molecular weights divided by the molecular weight of one of them. In this example, a mass accuracy of 0.00053 or 530 parts-per-million (ppm) is required. The division of the difference by two arises since if either molecular weight is in error by greater than one-half of the difference, the measurement will not enable the two elemental compositions to be distinguished. Such determinations are not limited to small molecular weight compounds provided that sufficient mass accuracies can be obtained. It has recently been shown using an ion cyclotron resonance mass spectrometer that two peptides with molecular weights that differ by just 0.00045 Da could be distinguished. This mass difference corresponds to a value less than the mass of a single electron (Figure 1.4).

1.3.2 Nitrogen Rule The nitrogen rule is of use in assigning an elemental composition to a compound based upon its molecular weight. For any compound that contains only C, H, N, 0, S, Si or halogen atoms (F, C1, Br or I), the nitrogen rule states that the nominal molecular weight for the compound will be even only if the number of nitrogen atoms is an even number or zero. The nominal molecular weight represents only the integer portion of the value. In the case of ethanol, the nominal molecular weight is 46. This is an even number consistent with the lack of nitrogen atoms in the molecule. The value is inconsistent with the one nitrogen atom present in NO2. Thus based on the nitrogen rule, we can assign the correct elemental composition of C2H,0 even where a mass accuracy of less than 530 ppm is unattained. A compound containing an odd number of nitrogen atoms in addition to atoms of any of those elements detailed above, will have a odd nominal molecular weight. In the case of NO,, the nominal molecular weight is seen to be 45.

8

Chapter 1

RVMRGMR vs. RSHRGHR (SZH,

VS. N4O)

m(e-) = 0.00055 Da

906.491

906.488

906.494

Mass (Da) Figure 1.4 ESI FT-ICR mass spectrum of (M+2HI2’ ions of two peptides that difSer in mass by 0.00045 Da. The mass resolution at FWHM is calculated to be 3.3 million (Source: Fig. 2, Fei He, Christopher L. Hendrickson and Alan G. Marshall, Anal. Chem., 2001,73 (3), 647-650)

1.3.3 Double-Bond Equivalents An additional rule of use in assigning a correct elemental composition is the double-bond equivazents (DBE) or number of “double bonds plus rings” rule. Put simply, this rule calculates the number of double bonds or aromatic rings in an organic molecule with the elemental composition C,H,,N,O,. The number of DBE in such a molecule is defined by equation 1.3. DBE = x - y/2 + 2/2 + 1

(1.3)

In the case of ethanol described above, the DBE calculated based on the elemental composition C,H,O equals zero, a value consistent with only single bonds being present in the molecule. For a molecule of benzene with an elemental composition of C6H6 the DBE equals 4 [or 6-(6/2) + 13. This is confirmed by its structure since three double bonds are associated with the aromatic ring and a further DBE is associated with the ring itself.

Mass Spectrometry’s Beginnings

9

FURTHER READING J.J. Thomson, Recollections and Reflections, G. Bell and Sons Ltd., London, 1936. EW. Aston, Muss Spectra and Isotopes, Edward Arnold and Co., London, 1933. J.L. Putman, Isotopes, Penguin Books, 1960. R.B. Firestone, S.Y.F. Chu and C.M. Baglin, Table of Isotopes, 8th edition, John Wiley and Sons, New York, 1998.

CHAPTER 2

The Mass Spectrum 2.1 CONCEPT OF CHARGE AND THE MOLECULAR ION To this point, only the molecular weight of a molecule has been considered to contribute to its appearance in a mass spectrum. However, as recognized by Thomson, the particles detected in mass spectrometry are ions. Thus they have both mass and charge, the latter of which is important to their detection. Mass spectrometers are unable to detect neutral molecules and radicals; a charge must be imparted onto an atom or molecule before it can be studied. The reason for this is that charged molecules can be “handled” or their paths controlled through the use of electric and magnetic fields, while radicals and neutral molecules are unresponsive. Depending on the nature of the ionisation process, and also the nature of the atoms and molecules themselves, different ion types can be formed. The most common and traditional way in which ions are produced in a mass spectrometer is through the loss of an electron (equation 2.1). This often occurs by the initial collision of a gaseous atom or molecule with an electron in a process known as electron impact or electron ionisation (EI) (Chapter 3).

The product ion formed is known as a radical cation as it is an odd electron species with a positive charge. Since the product, M+’, has the same mass as the molecular weight of the compound M from which it was produced, it is known as a molecular ion. In general, the notation +* adjacent to a molecule’s structure indicates that the molecule is deficient of an electron without designating the site of the charge. The charge of the ion is equal to the charge of an electron (e) defined Coulombs (C). Were the ion to possess two charges, as 1.6 x perhaps through loss of electrons from two atoms or groups of atoms within the same molecule, the charge on the ion would be 2e. Thus the 10

The Mass Spectrum

11

charge of an ion is always some multiple of the charge of an electron or ze, often just denoted z. In a mass spectrum, ions appear at a massto-charge ratio defined mlz, where m is the mass of the ion and z is the charge. Since z often (though not always) has a value of one, early references in mass spectrometry refer to an ion's mass-to-charge ratio as mle. It is important to note that not all ions are formed by electron loss. Some electronegative atoms or molecules can attract an electron during electron impact (equation 2.2). These ions, denoted M-*, are still referred to as molecular ions. The same is true of even electron species formed by the adduction or loss of a charge carrying atom or group. An ion may be formed for instance by the protonation of a molecule. [M +HI' is also referred to as a molecular ion, or more strictly a quasi or pseudo-molecular ion since the mass of the ion is now larger than the molecular weight of the molecule by one atom of hydrogen.

In general terms, most ions formed, dissociated and studied in mass spectrometry are positively-charged ones. This is because their production is usually more efficient for most classes of compounds over their negatively-charged counterparts. This is not to say that negative ion mass spectrometry is not important or not used; indeed many important observations and applications are based on these experiments (see for example Chapter 7). The polarity of the lenses used to repel ions from the source is the same as that of the ions themselves. The polarity of the voltage applied to the detector, on the other hand, is opposite to that of the ions in order to attract them to the ion detector. This leads to the detection of either positive or negative ions, but not both simultaneously.

2.2 FRAGMENT IONS 2.2.1 Formation of Fragment Ions If sufficient energy is deposited into the molecule during ionisation, the molecular ions may dissociate into smaller mass fragments that themselves may be ions. For example, an odd-electron radical cation, M+', may dissociate to form two fragments one of which is an even electron fragment of product ion, and the other is a radical R' (equation 2.3). M" -+ F'

+ R'

(2-3)

12

Chapter 2

Alternatively, M+'could dissociate to produce a smaller mass fragment and a neutral molecule N (equation 2.4).

Only the ionic products F' and F" are passed through the instrument and detected. Beyond simple bond cleavages, fragment ions can be produced following the rearrangement of atoms if sufficient energy is available to facilitate bond cleavage and formation. Hydrogen atoms and protons, for instance, are frequently transferred from a remote site to the ionic centre prior to cleavage of the molecular ion. The most abundant ion peak in a mass spectrum, whether it is that of the molecular ion or that of a fragment, is referred to as the basepeak. Ion abundances are measured relative to the intensity of the base peak, set arbitrarily to loo%, and usually plotted to the top of the y-axis (Figure 2.1). F+or F+' /

t

100

base peak

reIative abundance 50

M+'

YO

0

-I

I

I I I

Figure 2.1 Representation of an EI mass spectrum for molecule M

2.2.2 Stability of Fragment Ions The relative stability of the fragment or product ion is a major factor that influences the appearance of fragment ions in a mass spectrum. For a fragment ion to appear in a mass spectrum, it must be produced from a significant proportion of the precursor ions and be relatively stable to further fragmentation. The most predominant fragmentation pathways are those with the lowest energy barriers to product and with the lowest change in free energy (AG, where AG is associated with changes in enthalpy, AH, and entropy, AS, according to AG = A H - TAS). Other factors, including the strengths of the bonds broken during the fragmentation process and the time allowed for dissociation, also influence the appearance of fragment ions in the mass spectrum.

The Mass Spectrum

13

A few factors that can stabilise a fragment ion are now considered. Although illustrated for simple organic ions, these effects can be extended to other classes of compound.

2.2.3 Stabilising Effects One of the most common ways that positive ions can be stabilised is through electron donation from a neighbouring atom or group of atoms. The transfer of electron density toward a positive centre is known as the inductive eflect. Alkyl groups in hydrocarbons have the ability to donate electrons to a electron-deficient positive ion centre, such that the stability of carbocations follows the order: (CH,),C+ > (CH,),CH+ > CH,CH2+ >> CH,’. Thus the fragmentation of hydrocarbons gives rise to mass spectra in which ions at mlz 57 ((CH,),C’) and 43 ((CH,),CH’) are more abundant than those at mlz 29 (CH,CH,+) and 15 (CH,’). In contrast, electron-withdrawing atoms (F, C1, Br, I) or groups (-OH, -NO,) have a destabilising effect on a neighbouring positive ion centre. A second stabilising effect is the mesomeric efiect. Here a positive ion centre is stabilised by its conjugation with multiple (unsaturated) bonds. Hence the ion is stabilised through delocalisation of the charge across the molecule or fragment. As an example, the ion CH, = CH-CH,+ (mlz 31) can be stabilised by charge delocalisation to the form ‘CH,-CH = CH,. This delocalisation of charge through bond resonance also stabilises a phenyl ion (equation 2.5). Ions at mlz 77 (C,H,’) are a signature of aromatic compounds in EI mass spectra.

+

2.2.4 Quasi-Equilibrium Theory The unimolecular decomposition of molecular ions into fragments can be explained by the Quasi-Equilibrium Theory (QET). The QET provides a theoretical description of how these processes take place inside a mass spectrometer. The ionisation of a molecule by electron loss (equation 2.1) or electron capture (equation 2.2) occurs within approximately lo-%, a time much shorter than that required for a molecular vibration. Hence the ionisation event can be considered to be a “vertical transition” with no change in internuclear distances (Figure 2.2). If the geometry of molecular ion differs from that of the neutral molecule,

14

Chapter 2

t

energy

reaction coordinate

-

Figure 2.2 Schematic representation oj'energy levels associated with the ionisation and dissociation of a polyatomic molecule A4 along a reaction coordinate

the states for the latter will appear slightly along the reaction co-ordinate (to products) and the transition is said to be non-vertical or lddiabatic. Since pressures within the ion source are typically very low, each ion can be considered to be an isolated system as ions do not collide with each other or background gas molecules. If the excess energy transferred to a molecule during the ionisation process is distributed between all possible excited states in the ion, and these excited states interconvert between one another, a quasi-equilibrium is said to exist among these states. Therefore it is postulated that each excited state is of equal probability and that an ion's dissociation pathway depends only on its internal energy and the structure of the molecule, and not upon the initial site of ionisation or the nature of the ionisation process. Few exceptions have been found to the QET and such dissociations are termed non-ergodic. The ionisation energy (IE) is defined as the minimum energy necessary to produce a molecular ion, M" from its ground neutral state, M. It can be determined by raising the energy of the electrons in an electron impact source until a molecular ion is detected. Since the energy is related

The Mass Spectrum

15

directly to the potential through which the electrons are accelerated, it is often referred to as the ionisation potential. The appearance of a fragment ion in a mass spectrum occurs if there is sufficient energy to ionise the neutral molecule and overcome the activation barrier. This is referred to as the appearance energy (E) of the fragment ion. An additional factor, the kinetic shift, may contribute to the appearance of a fragment ion and results from the excess energy required in order for the fragmentation process to occur within the time the molecular ion spends in the ion source. The kinetic shift increases with the size of the molecule, the activation energy and the “tightness” of the transition state. Where there is no activation barrier and the kinetic shift is zero or negligible, the minimum energy required for (M) to dissociate into fragments can be defined as IE. All molecular ions with internal energies less than E, do not dissociate regardless of the time available for fragmentation. The appearance energy of a fragment can be defined by equation 2.6.

In the case in which two dissociation pathways are possible where (E,), =: (E&, the pathway which proceeds through a “loose” transition state, [M”]*, will predominate. This is because the energies of the excited states of a transition state in which the components are loosely associated will be similar to one another, over those for a “tightly-activated’’ transition state, such that they are easier to surmount. Ion dissociations involving the breaking of a single bond (with a bond dissociation energy of Do) usually proceed via a loose transition state complex and are typically favoured over rearrangement reactions that involve tighter transition states. Computer algorithms are available to calculate the energies of such states for low values of ( E Eo).Difficulties, however, arise for a real molecular system in predicting an ion’s physical parameters, including its energy for activation. Both theoretical and experimental results have shown that the rate constant k for a fragmentation process increases proportionately with the internal energy of the ion before reaching a maximum plateau beyond which no rate enhancement is observed. The shape of the curve from a plot of k versus energy E is predominately determined by the geometry of the transition state and the value of E,. Use of the QET allows the rate of fragmentation of an ion with a given internal energy to be predicted. The maximum rate constant for a simple bond cleavage process is of the order of 1014s-’. -

16

Chapter 2

A simplified version of the QET enables the abundances of molecular and fragment ions in mass spectra to be described in a semi-quantitative manner without the aid of a computer. In this version, the rate of fragmentation of an ion ( k ) is given by equation 2.7 where v is a frequency factor influenced by the entropy of the process and N the number of oscillations (rotations or vibrations) possible. k(E) = v[(E- E0)/qN--'

As explained above, the appearance of fragment ions in a mass spectrum is ultimately influenced by the time allowed for such fragmentation processes. Ions spend approximately 10% (or one microsecond) in an electron ionisation source so that only relatively fast fragmentation processes occur in this region of the mass spectrometer. On some mass spectrometers, an entire mass spectrum is recorded in just a few hundred microseconds providing energetically-excited ions with only this amount of total time to fragment. 2.2.5 Metastable Ions Metastable ions, denoted m*, are those formed by unimolecular dissociation of molecular ions in the field-free regions of the mass spectrometer anywhere between the ion source and detector. They can be useful in establishing the fragmentation pathways of molecular ions by unequivocally linking a fragment ion with a specific precursor. Consider a molecular ion of mass mp that dissociates to a fragment ion plus a radical or neutral molecule (equations 2.3 and 2.4). If a metastable fragment ion is produced in flight from the ion source to the detector it will have less translational energy than a comparable fragment ion generated within the ion source (of mass mf). This results in the ions appearing in the mass spectrum at an apparent mass (m*) that is less than that of fragments generated in the source (m,). On a magnetic sector mass spectrometer, the mass of a metastable ion, m*, is expressed according to equation 2.8.

The widths of the ion signals of these metastable ions of mass m* are considerably greater than those of other ion fragments. This is because the ions are not energy focused as they leave the ion source and the kinetic energy of the precursor ions is released isotropically as they are formed. Thus metastable ions have a broader range of energies over other fragment ions.

The Muss Spectrum

17

As an example, the loss of a methyl radical from ionised propane (of a molecular weight of 44 Da) gives rise to a fragment ion at mlz 29 and a metastable ion at mlz 19.1 (292/44= 19.1) Because the metastable ions appear as broad peaks in the mass spectrum, their mass-to-charge ratios are usually quoted to only a few decimal points at most. 2.3 RELATIVE ION ABUNDANCE The appearance of an ion in a mass spectrum is the result of an electrical current that is generated and amplified when the ion strikes the detector. A measure of the ion current across all ions in a mass spectrum is referred to as a total ion current (TIC). The charge on a singly charged coulombs (C). Since one ampere represents one coulomb ion is 1.6 x of charge per second, when one million singly-charged ions strike the detector a current of 1.6 x amperes (A) will be produced. The vertical or y-axis of a mass spectrum is usually plotted to display the ion current, the number of ions, or more commonly relative ion abundances. Here the ion signal with the highest current is normalized to 100% on the relative abundance scale and all other ions have abundances measured relative to this peak, the base peak. The relative abundance of an ion is dependent on a number of factors including the stability of the ion, the stability of the neutral product (in the case of a fragment ion from a unimolecular decomposition), suppression effects, mass resolution, and detector efficiency. Ion detectors do not detect ions across the mlz scale with equal efficiency, and it is common for ions at high mlz to be detected less readily than those at low mlz. Mass resolution also impacts ion abundance measurements. An ion signal associated with two unresolved ions would have a higher intensity reflecting the contribution of both ion currents to the signal than would be the case if each were mass resolved. Suppression effects have also been observed widely in mass spectrometry. In some cases, a neighbouring ion can completely suppress the signal of another ion such that it is not detected. When ionised separately, both ions are easily detected. Entropy effects are an important consideration in terms of the intensity or abundance of fragment ions formed by unimolecular dissociation. Entropy considerations favour the production of fragment ions by simple cleavage reactions over rearrangements involving a change in the molecular structure.

18

Chapter 2

2.4 MASS RESOLUTION The ability to separate two ion signals from one another in a mass spectrum is defined by mass resolution (R). Traditionally, this measure has been made based on the resolution of two ion signals above 10Y0 of their height (the so-called 10% valley definition) (Figure 2.3). Mass resolution is defined by equation 2.9 where M , represents the mlz ratio for the first ion and M2 represents the rnlz ratio of the second, and M , > M2.

Therefore in order to resolve ions at mlz 1000 and 1001 with a 10% valley, a mass resolution of 1000 is required. The same is true for ions at mlz 500 and 500.5. In some cases, a mass resolution is referred to as a ratio, namely 1:1000. Mass resolutions vary due to a number of factors including the initial kinetic energy of the ions as they exit the ion source. Most ions of the same mlz are formed with a range of initial energies (often represented by a bell-shape distribution), a result of their proximity to or remoteness from the acceleration lenses to which high potentials are applied. This may be corrected for by the mass analyser by what is known as energy focusing. Here only ions of the same mlz and energy are allowed to pass to the detector but this is achieved at the expense of ion detection. In cases where this energy spread is not corrected for, the width of the ion signal recorded at the detector will be larger.

100-

740

10

AM

at 10% valley

Figure 2.3 Dejinition ojmass resolution at u 10% vulley and full-width at huljmuximum (FWHM)

The Mass Spectrum

19

The most important factor, however, that impacts mass resolution is the type of mass analyser used. Magnet-based instruments as a rule achieve superior mass resolution to quadrupole and time-of-flight mass analysers. However, a better understanding of ion motion and optics inside a mass spectrometer has led to improvements in instrument design that have considerably increased the mass resolutions that can be achieved on most instruments. Mass resolutions of the order of 100-500 are considered low, those of the order of 1000-5000 are considered moderate, while those above 10,000 are considered to be high (see Appendix 4). Fourier transform ion cyclotron resonance (FT-ICR) mass spectrometers, described in the next chapter, offer the highest mass resolutions attainable to date, routinely up to 100,000. The ion signals associated with isotopes, even for large macromolecules, can thus be resolved to baseline (or the x-axis) on these instruments. Where isotopes or components in mixtures are not mass resolved, some measurement of mass resolution is often desired. In this case, it is common to measure mass resolution based upon the width of the ion signal at half its height. This measure is referred as full width at halfmaximum (FWHM) (Figure 2.2). For an ion at rnlz 1000, with a peak width of 2u at half its height, the mass resolution is said to be 500 (or 10OOl2).

2.5 MASS MEASUREMENT AND ACCURACY The capacity of mass spectrometers to resolve ions associated with heavy isotopes (e.g. 13C,'H) from those that contain only the lightest isotopes, enables a mass measurement to be made based on any one or several of these ion signals. For small molecules, the ion signals associated with heavy isotopes are typically of a low relative abundance since such isotopes are rare in nature. As a result, most mass measurements are based on the peak of Zightest isotopes, and the resulting molecular weight of the compound is referred to as its monoisotopic mass. For carbonbased compounds, this is also referred to as the "C-only muss, since ion signals at the lowest mlz ratios in the isotopic distribution contain no I3C or I4C. Carbon is considered exclusively here since the natural occurrence of heavy isotopes for other common elements in organic compounds are much lower compared with carbon. Where an ion's isotopes are unresolved, a molecular weight measurement is based upon the rnlz value at the peak top or a centroid value (where the centre of the peak top is determined from the area above a particular ion intensity level). Such a measurement yields an average mass, which contains a contribution to the molecular weight of the

Chapter 2

20

compound from all elemental isotopes. By definition the average molecular weight of a compound is always greater than its monoisotopic value. This difference between the monoisotopic and average value can be quite significant for large molecules. As an example, the monoisotopic molecular weight of the protein ubiquitin, with an elemental formula of C378H629Nlo,0,18S1 has a calculated value of 8,559.62 and average value of 8,564.86 Da. Special care must be taken where mass measurements are made based upon the calibration of the mass-tocharge scale using some ion signals with isotopes resolved and others where they are not. As the molecular weight of a compound becomes large, the probability that any of its ions contain no heavy isotopes becomes exceedingly small. As a result, the peak of lightest isotopes for any of its ions becomes small relative to those ions that contain heavy isotopes. It is necessary in these circumstances to base the molecular weight on the mlz value of ions containing some level of heavy isotope. In the case of the electrospray mass spectrum of bovine ubiquitin, a mass measurement based on the ion signal for ions containing five I3C atoms (or some other combination of heavy isotopes of equal mass, 4.g. I3C4+ 2H) at mlz 779.613, is 8,564.66 (11 x 779.613 11 x 1.0078 to adjust for the 11 protons attached to the protein that gives the ions their charge). This experimental value is within 0.03 u of the theoretical value (8,564.63) at this level of isotope enrichment. Hence a mass accuracy of 0.04/8,564.63 = 3.5 x or 3.5 parts-per-million (ppm) has been achieved. -

100.0

(M+llIifff mh 779.613

Avnage ResolviogP o w 0 200 :O00

1 "C,

0.0

mh779.613

,

80b.O

850.0

9Ob.O

950.0

lodo.0

lOh.0

Ildo.0

d X

Figure 2.4

ESI mass spectrurn of the protein bovine ubiquitin recorded at high mass resolution (-200,000 at FWHM). Insert shows resolved isotope peaks for the [ M + IIH]"+ ions.

The Mass Spectrum

21

The difference in mass-to-charge ratio between the isotope peaks for any ion (shown enlarged for the [M + llH]"+ ions to be 0.09 or 1/11 in Figure 2.4) is llz. Thus the charge on any ion can be determined directly from the mass spectrum where its isotopes are mass resolved. The level of heavy isotope content must be known or predicted in order for the molecular weight of the compound not to be in error. Alternatively, the compound can be constructed in an environment depleted of heavy isotopes such that ions associated with only the lightest isotopes for all elements are detected.

FURTHER READING F.W. McLafferty and F. Turecek, Interpretation of Mass Spectra, 4th edition, University Science Books, 1993. E. DeHoffmann and J. Charette, Mass Spectrometry - Principles and Applications, 2nd edition, John Wiley and Sons, New York, 2001. H.M. Rosenstock, M.B. Wallenstein, A.L. Wahrhaftig and H. Eyring, Absolute Rate Theory for Isolated Systems and the Mass Spectra of Polyatomic Systems, Proc. Natl. Acad. Sci. U S A . 1952, 38, 667-678. R.G. Cooks, J.H. Beynon, R.M. Caprioli and G.R. Lester, Metastable Ions, Elsevier, Amsterdam, 1973.

CHAPTER 3

The Mass Spectrometer 3.1 BASIC COMPONENTS All mass spectrometers consist of three basic components, the: (i) ion source, (ii) muss anulyser, (iii) ion detector. The role of the ion source is to introduce molecules into the mass spectrometer and convert them to a charged or ionised form. The ion source like the rest of the mass spectrometer is usually, though not always, held at a low pressure. Mass spectrometers are operated under vacuum to prevent the collision of ions with residual gas molecules during their flight from the ion source to the detector. This is because the ions are formed with excess energy and this, together with their charged character, can result in their reaction with other gaseous material present. To avoid this, the levels of contaminants and atmospheric gases such as oxygen within the ion source should be minimised. The ideal operating pressure is that in which the average distance an ion travels before colliding with a gas molecule (its mean freeputh) is longer than the distance from the source to the detector. After ions are formed in the source, they are accelerated into the mass unalyser where they are separated in vacuo according to their mass and charge through the use of electric and/or magnetic fields. Finally, the ions are passed onto an ion detector producing an electrical current that is amplified and detected. In most mass spectrometers, these three basic components are physically discrete entities. Thus each of them will be considered separately in order to understand how ions are formed, separated and detected in mass spectrometry experiments. It is not possible to review all the ionisation techniques that have been used in mass spectrometry experiments, some of which have been replaced by other more efficient methods. 22

The Mass Spectrometer

23

Instead, the following section discusses those most widely used approaches either alone, or in conjunction with, chromatographic and electrophoretic separations. 3.2 IONISATION TECHNIQUES AND INTERFACES

3.2.1 Electron Ionisation

As described in Section 2.1, the traditional method of ion production in mass spectrometry is electron impact or electron ionisation (El) in which gaseous sample molecules are bombarded with a stream of electrons (equation 2.1). Other processes by which ions can be formed during electron ionisation include dissociative ionisation (equation 3. l), ion puir formation (equation 3.2) and electron capture in which negatively charged ions are produced (equation 2.2). The electron ionisation technique is widely used for the study of relatively volatile organic molecules by mass spectrometry. AB + e- -+ A" + B + 2e-

(3.1)

AB + e- 4 A'+ B-+ e-

(3.2)

The electrons are produced by heating and passing a current through a thin ribbon of metal (such as ruthenium) known as the filament or cathode. The electrons are projected across the ion source by their attraction to an anode on the opposite side of the chamber (Figure 3.1). The energy of the electrons depends on the difference in the potentials applied to the cathode and anode. If a voltage difference of 70V is maintained, the electrons have energies of 70 electron volts (ev), or the equivalent of 6.8 x lo3kJ mol-'. A small magnetic field is usually applied across the ion source to cause the electrons to follow a helical path in order to increase the probability that they interact with a gaseous sample Torr or 1.3 x Pa is maintained molecule. A pressure of typically in the ion source. The collision of electrons with the sample molecules, M, often leads to their ionisation by electron loss with the formation of the molecular ion, M" (equation 2.1). The energy required to ionise a molecule depends on the molecule itself but, as an example, an energy of 10.5eV is needed to ionise a molecule of ethanol. Most organic molecules ionise in the range of 8-15 eV. Yet since not all the electron energy is necessarily transferred to the sample molecules during collision, higher electron energies of -70 eV are commonly used in EI mass spectrometry.

24

Chapter 3 repeller

filament anode

1 mass analyser

Figure 3.1 Schematic representation of an electron ionisation source

Regardless of the energy of the electrons, the sample molecules will possess a range of energies up to and including the energy of the electrons. Therefore some molecules will have sufficient energy to ionise (equations 2.1 and 2.2), others will not, and still others will contain enough energy to dissociate (equations 2.3, 2.4, 3.1, 3.2). Several factors contribute to the formation of fragment ions including the strength of the bonds to be broken, the stability of the products of fragmentation (both the ions and neutrals or radicals), and the internal energy of the fragment ions themselves. Where the dissociation of the molecular ions is problematic (and as shown in Chapter 5 it is often useful to determine the structure of a molecule), a lower potential difference (10-1 5 V) can be applied between the filament and the anode. The voltage difference at which molecular ions are first observed in a mass spectrometer is known as the ionisation potential. A few volts above this, molecular ions are mostly formed. As the potential increases further, more fragment ions are produced (Figure 3.2). In principle, mass spectrometers can be used to measure ionisation potentials. Because these measurements can be unreliable, a better measure is the efficiency of producing fragments. The appearance potential is related to the overall energy for the processes shown in equations 2.1, 2.3 and 2.4. This information is useful for measuring the thermodynamic parameters of a molecule such as its heat of formation and bond dissociation energies (see Chapter 6).

25

The Mass Spectrometer

-1

F+ only

t

log (ion

0

30

-

60 potential (V)

90

electron energy (eV) Figure 3.2 Plot of ion current versus the potential diference between the$larnent (cathode) and anode. The ionisation potential represents the voltage at which molecular ions are-firstdetected

Since electron ionisation can lead to the production of fragments (F') (equation 2.3) as well as intact molecular ions, it is referred to as a hard ionisation method. The remaining ionisation methods discussed in this chapter result predominantly in the formation of ions without fragmentation, and hence are known as soft ionisation techniques.

3.2.2 Chemical Ionisation Chemical ionisation (CI) is related to the electron impact method except that ionisation of a reagent gas, rather than the sample molecule itself, occurs first. This is followed by the transfer of charge to the sample molecule by a chemical process. One of the most common reagent gases is methane. When subjected to electron impact, a molecule of methane can ionise to form CH4+'by electron loss. This ion can react with a second molecule of methane, to produce CH,' (equation 3.3). CH,+' + CH4-+CH,++ CH,'

(3.3)

The ion CH,' is an efficient proton donor, so that a sample molecule M also present in the ionisation chamber can be ionised according to equation 3.4. M + CH,f

-+[M +HI'+

CH,

(3*4)

26

Chapter 3

To prevent the direct ionisation of molecules M, methane is present in the ion source at a much higher concentration than the sample. Because of this, a chemical ionisation source operates at a much higher pressure to lop4Torr, or 0.1 to 1 Pa) than an EI source. ( There are many chemical processes other than proton transfer that can be achieved inside an ion source to effect chemical ionisation. These include charge transfer (or charge exchange) (equation 3 3,ion-molecule addition (equation 3.6), and even nucleophilic displacement reactions in the case of negatively-charged ions (equation 3.7). M"+N+M+N"

M'+ N

-+ [M + N]'

NU-+ AB +-NU-A+ B-

(3.5) (3.6)

(3.7)

3.2.3 Coupling Gas Chromatography to Mass Spectrometry (GC-MS) The ability to ionise volatile molecules within a mass spectrometer led to mass spectrometers being coupled to gas chromatographs (GC). In these experiments, the volatile components of relatively complex sample mixtures can be ionised (by either electron or chemical ionisation) and detected in a stepwise manner as they are released from the GCcolumn. Modern GC-MS mass spectrometers use capillary columns (with 100-500 pm internal diameters) that provide for the separation of low levels of analytes and also minimise the amounts of G C carrier gas that enter the ion source. This helps to maintain a low operating pressure in the source. Modern GC-MS interfaces are designed to further minimise the levels of carrier gas that enter the ion source by using high capacity pumping systems. However, it is a requirement that some compromise be made between optimal G C conditions and those required for MS operation. A so-called open-split interface in which a proportion of the eluent from the column is pumped away before it enters the ion source is one way to achieve the coupling of a G C and mass spectrometer. This interface also enables the GC column to be interchanged without breaking vacuum ("venting") to the mass spectrometer.

3.2.4 Field and Plasma Desorption Ionisation These two ionisation techniques have largely been superseded by other methods, but they are mentioned briefly here since they represent the first

The Mass Spectrometer

21

methods available for ionising non-volatile molecules. Field desorption (FD) ionisation is achieved by depositing the sample onto a metal filament coated with carbon. A potential difference is applied between the filament and a nearby electrode such that ions are desorbed from the surface. The method is particularly useful for large non-polar compounds such as hydrocarbons but requires some skill to correctly prepare the coated filament. In plasma desorption (PD) the sample is deposited onto a foil constructed of nickel or aluminium-coated nylon. The fission fragments from the radioactive decay of californium-252 then pass through the foil. Californium-252, an isotope with a half-life of about 2.6 years, is a very efficient neutron source with one microgram producing about 170 million neutrons every minute. The fission particles deposit considerable energy into the sample and lead to the direct release of ionised forms of the sample molecules, usually [M + H]+and [M + Na]’ ions. Plasma desorption (PD) ionisation was the first ionisation technique capable of ionising non-volatile sample molecules with molecular weights of the order of 10,000 Da including polar molecules such as proteins. However, the use of a radioactive source and difficulties with preparing the sample for ionisation led to it being largely replaced by fast atom bombardment (FAB) ionisation shortly after its discovery. 3.2.5 Fast Atom or Ion Bombardment Fast atom bombardment (FAB) is related to an approach widely used to study the chemical nature of materials and surfaces called secondary ion mass spectrometry (SIMS) (see Chapter 9). In SIMS, a primary beam of high-energy (typically 10-30 keV) ions such as Xe+ or Cs’ bombard a solid surface, releasing secondary sample ions for analysis. One problem with the approach is that once the sample is ablated from a position on the surface, the yield of secondary ions decreases unless the primary ions are made to strike a different position. To overcome this, Michael Barber and colleagues invented the FAB approach in which the sample compound was suspended or dissolved in a non-volatile viscous liquid. Once a portion of the sample is ablated or sputtered from the surface by a primary beam of atoms or ions, the liquid “matrix” flows back across this region restoring a proportion of sample (Figure 3.3). The matrix also serves an additional role by dissipating the energy from the primary beam to minimise molecular damage to the sample.

28

Chapter 3

primary atom / ions

I

0 OO

secondary

- anaiyte ions

- a

matrix ions

0

o o O -/ molecules liquid matrix redisperses to bombardment site

analyte + liquid matrix probe surface

Figure 3.3 Schematic representation of a fast atomlion bombardment (FAB) source

The original application of the FAB technique made use of highenergy atoms (rather than ions) of argon to bombard the sample dissolved in glycerol. These atoms are formed from a gas held at a relatively high pressure ( to lo-" Torr, or 0.1 to 1 Pa) in an atom gun by a charge exchange process with ionised gas (equations 3.8 and 3.9). Ar + e- -+ Ar" + 2e-

(3.8)

Ar" (fast) + Ar (slow) + Ar (fast) + Ar" (slow)

(3.9)

When the primary atoms or ions collide with the liquid surface, a charge is transferred to the involatile liquid matrix (M) molecules and subsequently to the analyte (A) sample molecules (equations 3.10-3.12). Ar"

+ M -+ M" + Ar

M"+ MH

MH'+ M'

MH'+ A -+ AH++ M

(3.10) (3.11) (3.12)

Thus secondary ions are produced from the liquid matrix (M+*and MH') as well as the sample analyte (usually AH'). Because the concentration of the former is much higher on the probe surface, FAB spectra have a characteristic high "background" at low to modest mlz ratios (up to 500 u) associated with ions of individual matrix molecules and their molecular clusters. This matrix background can obscure the detection of analyte ions below mlz 500 particularly where the levels of sample present are low. A representative early FAB mass spectrum for the peptide met-lys-bradykinin (MW 13I8 Da) is shown in Figure 3.4.

The Muss Spectrometer

29

Met-iys-Erodykinin (M+

ti)'

l3l9l

1273

130L

LL 1100

700

417

I

1200

800

1300

900

1000

Figure 3.4 FAB mass spectrum of the peptide met-lys-bradykinin (Source: M. Barber, R.S. Bordoli, G.J. Elliott, A.N. Tyler, J.C. Bill and B.N. Green. Fast atom bombardment (FAB) mass spectrometry: A new ion source for mass spectrometry, J Chem. Soc, Chem. Commun., 1981,325-321, Figure 1)

Several properties of the liquid matrix are important to the success of the FAB method. The matrix should be an involatile liquid in vacuum, it should be chemically inert, and preferably dissolve most analytes. FAB matrices often possess a low pKa to assist in the generation of positively-charged AH+ ions, or a high pKa where negativelycharged ions are to be produced. Glycerol, nitrobenzyl alcohol, thioglycerol and dimethylsulphoxide are common FAB matrices. Dithiothreitol and dithioerythritol can be added to the matrix thioglycerol (magic buZZet) to reduce disulphide bonds in proteins suspended or dissolved in the matrix. Ions are typically produced by FAB ionisation over a period of several minutes before it is necessary to replenish the liquid matrix and/or the sample. Primary ions such as Cs' have largely replaced high-energy argon atoms in most FAB experiments, and thus the experiments have been referred to as liquid SIMS (LSIMS) in some accounts. Samples can also be introduced directly from a high-pressure liquid chromatograph (HPLC) by means of continuous flow FAB. This technique involves adding the FAB matrix to the mobile liquid phase used in the HPLC experiment at approximately 5% by volume. The solution is pumped down the length of the FAB inlet probe onto a target (Figure

30

Chapter 3

TIator High Voltage

rrget

Filter

Injection Valve

y

r

ShaR to Fit Vacuum Lock

Pads

Seal

Capillary Position Adjust men1

Figure 3.5 Design of a continuousflow CF-FAB sample probe (Source: R.M. Caprioli and W.T. Moore, in Methods in Enzymology, McCloskey (ed), Academic Press, New York, 1990, Vol. 193, Ch. 9, p. 216, Figure 1)

3.5) or mesh frit. A filter below the probe tip is used to absorb excess matrix. Stable ion signals are produced when the amount of liquid delivered to the tip or frit is equal to the rate of evaporation. The reduced concentration of matrix on the tip over standard or static FAB leads to a lower background of matrix ions that can aid the detection of analyte ions. Although FAB ionisation is less practiced today due to the development of laser based and spray ionisation methods, it remains an important technique for many applications. It is still in use for the study of organic and smaller biological compounds in many laboratories.

3.2.6 Laser Desorption and MALDI Lasers operating in both the ultraviolet (UV) and infrared (IR) have been used to desorb and ionise samples from solid surfaces for some time. The transfer of energy from the laser pulse leads to electronic excitation of the sample. Laser powers vary from approximately 1O6 to 10’’ J sec-’ cm-’ where the total energy per pulse is of the order of a few millijoules to a joule. Apart from the region of the sample upon which the laser is focused, there is usually little excess energy dissipated through the analyte. That said, considerable decomposition of some analytes can occur following the laser pulse. Despite this, a number of moderatelysized (- 1000 Da) sample molecules such as oligosaccharides, peptides and polymers have been successfully ionised by laser desorption ionisation (LDI, or just LD). This “neat” desorption strategy was effectively replaced with the development of matrix-assisted laser desorption ionisu t ion (MA LDI).

31

The Mass Spectrometer

In MALDI, the analyte of interest is mixed with a large mole excess of (ca. 1,000-fold)a matrix compound that absorbs efficiently at the laser wavelength. The matrix allows the energy from the laser to be dissipated and also assists with the ionisation sample molecules through electron transfer and chemical processes. Both solid and liquid matrices have been used, though the former is by far the more successful in most applications. Where a UV laser is used, common MALDI matrices include nicotinic acid, 2,5-dihydroxybenzoic acid, sinapinic acid and a-cyano-4-hydroxycinnamic acid (Figure 3.6). It is important to note that all of the MALDI matrices in Figure 3.6 contain phenolic and/or carboxylic acid groups. It has been found that proton transfer from matrix to sample molecules is important in order to achieve efficient ionisation of many compounds and that this transfer occurs, at least in part, in the vapour phase above the sample plate. CN

I

Figure 3.6 MALDI matrices nicotinic acid, 2,5-dihydroxybenzoic acid, sinapinic ucid and a-cyano-4-hydroxycinnamic acid (in order from lejt to right)

The correct preparation and deposition of MALDI samples onto the sample plate is critical to the success of the method. Most sample solutions are diluted in a solution of matrix, and a small volume of the combined solution (1 p1 or less) is deposited onto the sample plate. The sample droplet is usually allowed to dry in ambient air (by so-called dried droplet evaporation). However, since the morphology of the crystallised sample surface affects the success of MALDI mass spectrometry, other methods to deposit samples have been developed. These include the addition of organic solvents to the solution, the use of heat to assist the drying process, and the electrostatic-spraying of solutions of analyte and matrix (either separately or in a combined form). The latter technique gives rise to extremely thin and uniform surface of both the analyte and matrix resulting in more reproducible mass spectra being obtained regardless of the position from which the sample is ablated by the laser. A further useful strategy that has been adopted where the concentrations of analyte are low is the deposition of droplets onto pre-coated sample surfaces to either localise (for example, through the use of

32

Chapter 3

Teflon-based surfaces) or immobilise the analyte molecules onto a very small sample area (4mm’). Subsequent chemistries can also be performed on the molecules on these surfaces prior to their analysis by mass spectrometry. One such MALDI-based approach has been dubbed SELDI for surface-enhanced laser desorption ionisation. Since the laser or sample surface can be easily repositioned during analysis, most MALDI-based mass spectrometers today make use of a sample stage onto which 100 to several hundred samples are loaded. The plates also resemble the size of a microgel or blot (some 10 cm2) to facilitate the direct transfer of samples (particularly proteins) after their separation by two-dimensional polyacrylamide gel electrophoresis (2D-PAGE). Greater success has been achieved in experiments where the proteins are blotted onto a membrane to which a solution of matrix is applied. Unfortunately, the direct ionisation of proteins from gels has proved more difficult due to the presence of detergents and other contaminants that impede the ionisation process. A representative MALDI mass spectrum for a protein mixture is shown in Figure 3.7. Consistent with most protein analyses, the dominate ions in the spectrum have the form [M + HI’ from which a molecular weight can easily be derived. Note in this case that insufficient mass resolution has been obtained in this case to resolve the isotopes for the [M + HI+ions. I

cc

[M+H]

[M+H]

i i 3 4 ii

+

+

12361

1

%

m/z

Figure 3.7 Linear MALDI-TOF mass spectrum o j t h e protein mixture, insulin ( I ) , cytochrome c ( C C ) and apomyoglobin ( A M )

MALDI is now a firmly established technique, particularly for the study of polar, high molecular weight compounds such as proteins, glycoconjugates and nucleic acids. In most mass spectrometry laboratories today, its use is complemented by the application of electrospray ionisation (ESI), a completely unrelated ionisation method described in the next section.

The Mass Spectrometer

33

3.2.7 Spray Ionisation Methods; Thermospray

One disadvantage of MALDI, and other ionisation methods, in which the sample is deposited in its solid state is the difficulty in performing high throughput separations in conjunction with mass spectrometric analysis. Although mass spectrometry can be used to analyse sample mixtures directly, some components may not be mass resolved and thus not be detected as the complexity of these mixtures increases. Various chromatographic and electrophoretic approaches are used widely for the separation of components within complex chemical and biological extracts. The development of solution-based spray ionisation approaches have enabled these technologies to be coupled directly to a mass spectrometer providing a further separation dimension prior to MS analysis. In 1983, Blakley and Vestal reported the development of the thermospray ionisation method for this purpose. Briefly, a dilute solution of an analyte is pumped through a stainless steel tube and heated to approximately 100 “C subject to the flow rate and nature of the solution. A jet of vapour containing a mist of solution droplets is projected into the ion source by a free jet expansion and preformed ions in solution are evaporated and detected. Thermospray ionisation has been applied to the study of small organic molecules and moderately-sized biological molecules such as peptides but has been largely superseded by electrospray ionisation. 3.2.8 Electrospray Ionisation

Electrospray ionisation (ESI) was first conceived in the late 1960s by Malcolm Dole and has developed from experiments performed in the late 1980s by John Fenn and colleagues. The electrostatic spraying of liquids is used in many industrial applications and involves passing a solution through a needle held at high voltage (typically 4-5 kV) relative to some counter electrode. When the solution is an electrolyte and the needle forms part of an ion source in a mass spectrometer, the fine mist of droplets that emerge from the needle tip possess a net positive or negative charge determined by the polarity of the needle and are attracted to the entrance of a mass analyser (Figure 3.8). The droplets emerge from what is known as a Taylor cone formed by the elongation of the electrolyte solution at the needle tip as like-charged ions are repelled from the needle. The application of a “counter-current” dry gas (that passes in the opposite direction to the passage of droplets) considerably aids droplet evaporation and it was this feature that led to the successful ionisation of large molecules.

34

Chapter 3 +V

-

+kV

analyte solution

00O @

7

t

@

@ @

Taylor cone

sheath gas

\

0

@

@

@

to mass

-

i

analyser

Figure 3.8 Magnified representation of the cross-section of an electrospray ion source.

As the droplets evaporate, the ions within them move closer together. At some point there are sufficient Coulombic repulsive forces between the ions to overcome the liquid surface tensions, resulting in the production of smaller droplets that continue to undergo the process. Eventually solvent-free ions are produced that are passed through the mass analyser and detected. A curious feature of the electrospray ionisation mass spectra of large biopolymers is that the ions produced are usually multiply-charged, with a continuous series of such ions being detected (Figure 3.9). The reasons for this are not entirely clear though the phenomenon has been linked to the time it takes for ions to emerge from the solution droplets during the evaporation process. 1

1

9+

8+

1o+

7+

i miz

Figure 3.9 Electrospray ionization ( E S I ) mass spectrum of protein chloroflexus thioredoxin exhibiting ions of the form [Mi-nH]"' (Source: K.M. Downard, Advances in protein analysis and sequencing by mass spectrometry, New Adv. Anal. Chem., 2002, P2,1-30, Figure 1 (adapted))

The Mass Spectrometer

35

An advantage of this feature is that a molecular weight measurement can be made based on each multiply-charged ion and the values averaged across all charge states. This leads to routine mass accuracies of k 0.01%, or a 1 Da error at a molecular weight of 10,000 Da even on mass spectrometers with modest mass resolving capabilities. To achieve this, the charge state for a particular ion must be determined. If it is assumed that two neighbouring ions (with mass-to-charge ratios of m,lzi and mjlzj where m,lz, < m,/zj) differ by one charge unit and both support the same charge-bearing species, then the charge on ion i ( 2 ; ) is given by equation 3.13 where mp is the mass of the charge-bearing ion, typically a proton. zi = (mj/zj - mp)/(mj/zj- m,/z,)

(3.13)

Once the charge of any ion is derived, a molecular weight measurement based on any ion signal can be determined from equation 3.14. MW = zi (m,/z,)- zi mp

(3.14)

Applying this equation in the case of the data presented in Figure 3.9, the charge of the protonated ions at mlz 1,335.4 is z = (1,502.2-1)’ (1,502.2- 1,335.4) = 9.0. The molecular weight of the protein based on this protonated ion signal is thus 9( 1,335.4)- 9( 1.O) = 12,009.6. The same molecular weight value is obtained based on the mlz for the [M + 8H]*+ ion; that is 8(1,502.2) - 8(1.0) = 12,009.6. Several computer algorithms have been developed to perform these calculations automatically and to display the output on a molecular weight scale in what has become known as a deconvoluted mass spectrum. One such algorithm considers the mass of the charge-bearing ions as a variable and constructs a three-dimensional deconvoluted mass spectrum from which the identity of the charge-bearing species can be determined and not assumed. 3.2.9 Atmospheric Pressure Chemical Ionisation

A related approach that is also capable of ionising polar molecules directly from solution is known as atmospheric pressure chemical ionisation (APCI). In this method, ionisation is achieved by an electrical discharge in the vicinity of gaseous sample molecules produced by vapourising the solution stream. Ionisation takes place through chemical processes such as those described in Section 3.2.2. The process is less

36

Chapter 3

efficient at ionising large molecular weight compounds, but does have utility for many modestly-sized polar biomolecules (to - 1000 Da). For this reason, the approach is employed widely in drug discovery investigations including the study of metabolites (see Chapter 8). It works well even for relatively high flow rates (several ml min-') that are common in analytical HPLC applications. Since the ESI and APCT ionisation approaches are related and have complementary applications, many instruments feature interchangeable ESI and APCI ion sources. A disadvantage of ESI mass spectra of large compounds is that many ions are associated with each component present in the sample solution. In the case of a complex sample mixture, these ion distributions could be incorrectly associated with one another so that molecular weight errors arise. However, to overcome this the ionisation method can be coupled to a liquid chromatograph or capillary electrophoresis system where some initial separation of components is effected prior to MS analysis. 3.2.10 Coupling Liquid Chromatography and Capillary Electrophoresis with Mass Spectrometry There are two major considerations in coupling liquid chromatography and capillary electrophoresis separation systems to a mass spectrometer. First, the solvent must be efficiently evaporated before ions leave the source to enable the pressure within the mass analyser to be maintained. This also aids the efficient sampling of ions. Second, the nature of the solvent and other dissolved components (buffers, salts, denaturants etc.) should not impede the ionisation process. There is clearly a compromise reached in order to optimise the separation of chemical and biological mixtures with their direct on-line detection by mass spectrometry. Nonetheless, reproducible data can be acquired when appropriate operating conditions are maintained. A typical liquid chromatographic ESI mass spectrometry (LC-ESIMS) apparatus is illustrated in Figure 3.10. In brief, solvent pumped from the reservoirs enters the injector and passes through the chromatography column to the mass spectrometer. A UV detector can also be incorporated such that components are detected by absorption spectroscopy and mass spectrometry either simultaneously or in tandem. Once the sample is injected into the loop, the solvent delivers it to the column where chromatographic separation of the components occurs. The components then pass one by one, or as simpler mixtures, into the ion source where they are ionised and ultimately detected. Note that since the mass spectrometer separates the ionic forms by mass analysis, it is not

The Mass Spectrometer

37

MS

t-

11 Pump 3

Figure 3.10 Representation of a liquid chromatographic ESI mass spectrometry (LC-ESIM S ) experiment

necessary that complete chromatographic separation of each component of a mixture be achieved. Thus the chromatographic separation step is less rigorous than that which would be required where no mass spectrometer was employed. In the case of liquid chromatography, solvent flow rates of the order of several nanolitres to millilitres per minute have been coupled to ESI and APCI sources. An approximately linear correlation between the yield of ions detected and the concentration of the analyte in solution is observed over several orders of magnitude. Thus a quantitative measure of the ion current for each component enables the relative concentration of the analytes in solution to be determined. The addition of standards of known concentration into the sample, or their analysis in a separate run, allows such LC-MS approaches to be used for quantitative applications. The preferred mobile phases are those that contain significant levels of a volatile organic solvent such as acetonitrile or methanol, both of which are widely used for chromatographic separations. Pure aqueous solvent systems, however, can also be managed albeit with slightly reduced performance. The presence of low levels of ion pairing agents such as trifluoroacetic acid can assist with generating preformed ions of the

38

Chapter 3

sample components in solution. High levels of salt and other buffers and denaturants, however, should be diverted away from the ion source, usually during the early stages of the run, in order to prevent their build up in the transfer lines and on the spray needle. One further issue in performing such experiments on most mass spectrometers is whether or not all ions across the mlz range of the mass analyser are detected simultaneously. Where mass analysers are used in which the electric andlor magnetic fields are scanned, ions produced from an analyte eluting over a relatively small time period may pass into the mass analyser but not be transmitted to the detector. To compensate (if not correct) for this, either fast scanning of the mass analyser is employed or the flow rate of the mobile phase is reduced. An alternate method to detect low levels of compounds in these experiments is to “park” the electric and/or magnetic fields of the mass analyser to transmit only ions of a particular mlz ratio onto the detector. This selected ion monitoring (SIM) mode is useful for quantitative analysis of particular components in chemical and biological samples. Capillary electrophoresis mass spectrometry (CE-MS) is achieved in much the same way as that outlined in Figure 3.10. Capillary electrophoresis separates charged compounds with high resolution and is compatible with ESI-MS. Problems with coupling capillary electrophoresis with a mass spectrometer stem from the relatively high levels of buffers and salts used that can clog the transfer lines and disrupt the ionisation process. Gel-filled capillaries can be utilised to remove buffers from the ion source. Alternatively, narrow bore (5-10 pm) capillaries are used to reduce the electrolyte flow. The use of low flow rates (nl min-’) in general offers improved sensitivities that are exploited in many applications of electrospray ionisation mass spectrometry.

3.2.1 1 Low Flow Rate Electrospray Ionisation - Nanospray Electrospray ionisation mass spectrometry is most often performed using solution flow rates of several plmin-’. There are several advantages, however, in using considerably lower flow rates (10-50 nl min-I). These include compatibility with micro-flow LC and CE separation, improved spray stability, and lower sample consumption. Such experiments can also be performed without the use of a sheath gas passing around the spray needle to direct the electrosprayed droplets. Since there is also little solvent to evaporate during the ionisation process,

39

The Mass Spectrometer

heating the source chamber and/or the use of counter-current dry gases can be avoided. These so-called micro-electrospray or nanospray experiments can also be performed off-line by spraying liquids from microfine capillaries prepared by etching or drawing out glass capillaries (Figure 3.11). The capillary is loaded with approximately 1 pl of analyte solution and the flow rate is maintained by the electrospray process without the need for a delivery device such as a syringe pump. The capillary is either coated with a conducting material or a non-corrosive conductive wire is passed through the capillary to supply the high voltage to the tip. A camera or microscope can be used to position the capillary a few millimetres from the entrance lens to the mass analyser.

1 rnm ID

1-3prn

1 mass analyser

1

Y

H

V cable

Figure 3.11 Nunospray needle is mounted on an adjustable support for positioning u few millimetres from the entvunce to the mass analyser

A comparison of the common ionisation techniques is presented in Appendix 3. Once molecules have been introduced into a mass spectrometer as their ions, a mass analyser is used to guide them to the detector through the application of electric and magnetic fields. 3.3 MASS ANALYSERS

Once ions have been formed and introduced into a mass spectrometer, a mass analyser is used to separate them based upon their mass-to-charge ratio through the application of electric and magnetic fields. There are a

40

Chapter 3

number of different mass analysers described in the following subsections. Many modern instruments feature several mass analysers coupled together for use in tandem mass spectrometry (see Chapter 4) and other applications. When a mass spectrometer is constructed of several mass analysers of a different type, it is referred to as a hybrid instrument. This section begins with the simplest mass analyser, the timeof-flight tube.

3.3.1 Time-of-Flight As the name implies, time-of-flight (TOF) mass spectrometers separate ions and measure their mlz based on the time they take to pass (“fly”) from the ion source to the detector. The flight tube is usually 1-2m in length and the basis of the separation makes no use of either electric or magnetic fields. Ions are separated in thefield-free region of the flight tube before reaching the detector. A simple representation of a TOF mass spectrometer is shown in Figure 3.12. ion source

flight tube

-

ion detector

distance, I

Figure 3.12 Representation of a time-of-flight ( TOF) mass spectrometer

Ions are first formed in the source and then “pushed” down the flight tube through the application of a high accelerating potential (V) of the same polarity of the ions applied to a lens or grid. The kinetic energy of ions of mass, m, and charge, z,is given by equation 3.15 where v is their velocity. All like-charged ions (common z ) share the same initial kinetic energy (KE) as they leave the ion source. KE = 1/2rnv2= zeV

(3.15)

The time, t, is takes for the ions to pass the length of the tube ( I ) is given by t = llv. Substituting for v in equation 3.15 and rearranging, leads to equation 3.16. (3.16)

The Muss Spectrometer

41

Since the length of the flight tube ( I ) and accelerating voltage ( V ) are fixed, the time it takes for the ions to reach the detector depends only on their mass and charge. If the time it takes for at least two ions of known mass-to-charge ratio to reach the detector is measured, the time scale can be correlated with mlz values. As a general rule of thumb, singly-charged ions of molecules of 10,000Da take about 100psec to reach the detector. Common accelerating voltages are of the order of 10-30 kV. As one might expect, a flight tube of a common length (1-2 m) would not be able to separate ions with very similar mass-to-charge ratios. An additional complication arises since, due to the spatial distribution of ions in the ion source and their proximity to the applied electric field, not all the ions receive the same initial kinetic energy. These factors give rise to relatively poor mass resolutions of the order of 100-500 in linear TOF mass spectra. This leads to components in mixtures being unresolved from one another and large errors (-1%) in molecular weight measurements. To overcome this, several features are now built into most TOF mass analysers that considerably improve mass resolution and thus mass accuracies. The first of these is an ion mirror, ion rejlector or simply a rejlectron (Figure 3.13). A reflectron is constructed of a stack of donutshaped lens connected by a series of resistors across which a high voltage (V,) is applied. In most instances the voltage difference between each lens of the stack is identical creating a linear or homogeneous field. Socalled curved or inhomogenous field reflectrons have also been developed and have some advantages for transmitting ions across a wide mlz range. The potentials applied across the lenses of the reflectron causes the ions that enter it to be gradually repelled. These ions are reflected down the same or second flight tube to a second detector. Improvements in mass resolution are achieved because ions of different kinetic energies ion source

flight tube

ion mirror / reflectron

Figure 3.13 Representation of a reflecting time-of-flight (rTOF) mass spectrometer

42

Chapter 3

penetrate the mirror to differing degrees. Furthermore, the reflectron effectively extends the flight tube to almost twice its length which (from equation 3.16) can be seen to have a dramatic effect on an ion's flight time. Note that the instrument can also be operated in a linear-mode since, where no voltage is applied to the reflectron, ions pass through it to the first detector. Consider for the moment two ions of the same mlz that have slightly different initial kinetic energies, KE, and KE, (and velocities v1 and v,) when they leave the ion source, where KEI > KE,. Ions with greater kinetic energy will pass further into the reflectron before being repelled, while ions with less kinetic energy will travel over a shorter distance. This difference in the flight path and time corrects for the differences in the kinetic energies of the ions so that they reach the detector at the same time. A dramatic improvement in mass resolution is evident in TOF mass analysers operating in the reflectron over the linear mode as is illustrated for a segment (residues 18-39) of the peptide adrenocorticotropic hormone (ACTH) (Figure 3.14). The best mass resolution is achieved when ions spend equal times in the reflectron and the flight tube of the reflecting time-of-flight (RTOF) analyser. Linear

L4bl 2464 2466 2468 2470 2472

Linear DE

2462 2464 2466 2468 2470 2472

Reflectron DE

2462 2464 2466 2468 2470 2412

Figure 3.14 TOF muss spectru of the [Mi-HI' ions of the peptide ACTH18-39 recorded in the linear (left), linear with delayed extruction ( D E ) (centre) and reflectron DE mode (right)

A second feature to improve mass resolution that is employed today in most TOF instruments is time-lag focusing (TLF). Time-lag focusing has been revisited in recent years to improve the performance of MALDITOF experiments and has been described by different names including delayed or pulsed-ion extraction (DE or PIE). In conventional experiments, ions are extracted from the ion source through the application of an accelerating potential immediately after they are formed. If a time

The Mass Spectrometer

43

delay is introduced before the application of this potential, ions formed with more initial kinetic energy and greater velocities will move further from the ion extraction lens or grid. Application of an accelerating potential pulse imparts more energy into the ions further from the lens than those closer to it. The amplitude is adjusted so that the initially lessenergetic ions further from the lens will catch up to the initially moreenergetic ions so that they all reach the detector at the same time (Figure 3.15). The improvement in mass resolution that can be attained is illustrated in Figure 3.14. ion source

-~

part of flight tube

I

I

*! 4 9 t-@

I

@+;

*; i h: I

f

0,

@Dt

: I -

t

Figure 3.15 Principle of time-lag focussing (TLF) on a time-of-ight mass spectrometer

Mass resolutions of up to 30,000 (at FWHM) can now be achieved on TOF mass spectrometers by making use of both time-lag focusing and ion reflectrons. This is a dramatic improvement over their historical mass resolving capabilities. Consequently, TOF instruments can be described as analysers that achieve modest to high mass-resolutions that are second only to those of magnetic-based instruments.

3.3.2 Magnetic Sector Instruments that contain a magnet positioned over one region of the ions’ flight path are the oldest type of mass spectrometer. An ion of charge z moving with a velocity, v, that transverses a magnetic field B at right angles to the direction of the field will experience a centrifugal force given by zevB. When this force is equal to the centripetal force, ions adopt a circular path of radius r (equation 3.17). zevB = (mv2)/r

(3.17)

When equation 3.17 is rearranged, equation 3.18 is produced. r = (mv)/zeB

(3.18)

44

Chapter 3

This equation indicates that for ions of a particular charge z moving through a fixed magnetic field, B, the radius of their path is dependent only upon their momentum mv.In other words, ions of the same charge will follow a different path when they move pass the magnet influenced only by their mass and velocity (Figure 3.16).

acceleration lens

ion

ion detector Figure 3.16 Pussuge cfions of diferent mlz through muss spectrometer

M

magnetic.field of u single.focusing

Since the initial kinetic energy of the ions 1/2rnv2equals ze I f , the initial velocity of the ions is dependent on the potential, V, through which they are accelerated. Rearranging for v2 we arrive at equation 3.19. v2 = (2ze V)Im

(3.19)

From equation 3.17 we can derive: v = (zeBr)lm

(3.20)

If we square both sides of equation 3.20 we get: v2 = (zeBr)21m2

(3.21)

If equations 3.19 and 3.21 are combined and rearranged, we arrive at equation 3.22. mlz = eB2r212V

(3.22)

Therefore specific values of V or B allow ions unique in mass-tocharge to pass through the magnetic field along a path to the detector. Variations in either Vor B will cause these same ions to follow a different trajectory and collide with the walls of the flight tube. In Figure 3.16, only ions following the centre trajectory reach the detector at any set of

45

The Mass Spectrometer

V and B values. In practice, a series of slits are used throughout the instrument to further improve the focusing and separation of ions. It follows that a complete mass spectrum, in which all ions in turn are passed to the detector, can be recorded by changing (scanning) Vor B over time. In practice, when the accelerating voltage is too low insufficient numbers of ions will leave the ion source and reach the detector. For this, and other reasons, scanning of the magnetic field B is preferred. However, the scan rate of this type of mass analyser is limited by hysteresis where the magnetic field can become perturbed. To minimise this, magnetics are scanned more slowly than other mass analysers with time allowed between scans to “settle” the field. Laminated magnetics, however, allow for more rapid scan rates, approaching 0. l sec decade-’ (where a decade is equal to a range covering an order of magnitude difference in mass units, e.g. 100 to lOOOu), to be achieved. As mentioned earlier, ions leave the source with a range of kinetic energies rather than a single value due to their spatial distribution. Since some ions of the same mass will have different velocities and will still reach the detector for a particular set of B and Vvalues, the mass resolution achieved by a single magnet is compromised. To minimise this problem, most modern magnetic sector mass spectrometers also feature an electrostatic or electric sector. If a radial electrostatic field E is created by two curved plates held at oppositely charged potentials (+E and -@, an ion of charge z moving with a velocity v will transverse the field when its electrostatic force equals the centripetal force (equation 3.23). zeE = (mv2)/r

(3.23)

Since the kinetic energy of an ion 1/2mv2equals zeV, equation 3.24 becomes: r = 2VlE

(3.24)

Note that the trajectory of the ion defined by Y is independent of its mass and charge. At a fixed accelerating voltage, V, the ion’s trajectory is thus dependent only on the electric field strength. Mass spectrometers that combine electric and magnetic sectors are known as double-focusing instruments. In most cases, the electric sector is positioned before the magnetic sector in terms of the direction that the ions travel (Figure 3.17). These instruments are referred to as forward geometry instruments. Mass spectrometers with the reverse order of sectors are termed reverse geometry. The order of mass analysers after

46

Chapter 3

1 = --

,-

acceleration lenses source slit

U

ion source

collector slit

9

ion detector

Figure 3.17 Pussuge of ions through u double focusing ( E B ) magnetic sector muss spectrometer

the ion source is often abbreviated simply as EB or BE. Mass spectrometers can be constructed of even more (three, four, five and even six!) sectors that have particular uses for tandem mass spectrometry (Chapter 4) and studies of ion chemistry (Chapter 6). These are denoted EBE, BEB, BEE, EBEB or BEBE etc. Double-focusing instruments can be scanned in a number of ways but a common scan is one in which the electric and magnetic fields are varied such that the ratio of the field strengths is always held constant (BIE = constant). These scans are known as linked scans that will be returned to later in Chapter 4 in the context of tandem mass spectrometry (MS/MS) experiments on magnetic sector mass spectrometers. Double-focusing sector mass spectrometers can achieve mass resolutions up to 100,000 that allows ions which share the same nominal mass but different exact mass to be resolved. Accurate mass measurements (see Section 5.1) can be obtained where an ion’s mass is measured to six decimal places (or a few parts-per-million (ppm)). This can be useful to identify the composition of an ion as discussed in Section 1.3.1. Reverse geometry sector mass spectrometers are also useful in massanalysed ion kinetic energy spectra (MIKES) experiments in which metastable decomposition products are detected (see Section 4.3.1).

The Mass Spectrometer

47

3.3.3 Quadrupoles The concept of the quadrupole mass analyser was first reported by Paul and Steinwedel in the 1950s. A quadrupole (Q) consists of four rods arranged in parallel where those opposite to one another are electrically connected (Figure 3.18). The quadrupole has a number of advantages over magnetic sector mass analysers including the low cost of construction, their compact size, and fast scanning capability. A voltage of opposite polarity (+/-v) is applied to adjacent rods consisting of a direct current (DC) component (denoted U) and a radiofrequency (RF) (denoted VRFcos(wt))component where w is the angular frequency of the RF field. Ions are accelerated out of the ion source along the z-axis between the rods. They experience forces in the x and y direction - ze(d Vldx) and -ze(d Vldy) that cause them to oscillate toward and away from the rods. When the oscillation becomes too large the ions strike the rods and do not reach the detector. ion detector

P

resonant ion

+U+VRFCOS( 0 t)

II

non-resonant ion

Figure 3.18 Quadrupole mass analyser showing ion oscillation under the in.uence ojthe variab1e.fields

For any given analyser, the radius of an imaginary cylinder that fits in the centre of the rod (ro) is constant as is the frequency of the R F field w. Two functions a and q define a stable trajectory for which ions

48

Chupter 3

do not collide with the rods across a range of values for U and VRF (equation 3.25 and 3.26). a; = -2a, = -4ze UIm2r,’w2

(3.25)

q_= -2q, = -2ze VRFIrn2r~w2

(3.26)

A plot of a versus q is known as a stability diagram where the regions below the curves for m, m2 and m, represent the values of a and q for which ions follow a stable ion trajectory to the detector (Figure 3.19). In principle, the quadrupole can be operated over a range of values of U and VRF such that the (a,q) coordinates are always below the curves. In practice, the voltages U and VRF are held at a fixed ratio to maximise the mass resolution that can be achieved. This leads to an operating region defined by a line with a slope of 2U/VRF.A complete mass spectrum is obtained by scanning the voltages U and VRFwhere this fixed ratio is maintained. Ideally the voltages U and VRFshould be held constant throughout the passage of ions of a particular m/z through the analyser. Scan rates of the order of 1000 u sec-’ are common and allow for the majority of ions of a particular m/z to reach the detect or.

0.30

operating line

t a

0.10

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

Figure 3.19 Stability diagram indicating the values of a and q for which ions ofmass m,,m, and m3follow u stable trajectory to the ion detector

49

The Mass Spectrometer

Because quadrupoles operate at lower voltages and can be scanned at faster rates than magnet-based mass spectrometers, they are more easily coupled to gas and liquid chromatography instruments. However, they achieve lower mass resolutions of up to approximately 5000.

3.3.4 Quadrupole Ion Trap There are strictly two types of ion traps though the term is usually associated with only one of them: the quadrupole ion trap. Quadrupole ion traps (QIT or IT) are so named because they use similar operating principles to those of the standard quadrupole mass analyser. Despite this they are constructed very differently and consist of two conical lens or electrodes, and one “donut-shaped” ring lens (Figure 3.20). Ion cyclotron resonance (ICR) mass spectrometers are also ion traps but use magnetic fields to store ions as described in the next section. ring electrode

ion source

X

I

9

ion detector

Figure 3.20 Cross-section of a quadrupole ion trap

In a QIT, ions are held or trapped in the small interior volume between the conical lenses that form the “caps” of the trap and the centre of the ring electrode. By lowering and raising the voltages on the entrance and exit trap electrode, ions can pass into the trap, be stored for some period of time (usually ps), and then be released to the detector. The trap is usually operated in the mass selective stability mode where ions of a particular mlz are selectively released from the trap. Within the trap ions undergo a complex sinusoidal motion with the application of an oscillating RF potential to the ring electrode. An ion will be stored in the trap depending upon the values for the mass, m, and charge, z , of the ion, the radius of the ring electrode (yo), the separation of the caps from the centre of the ion trap (zo),the oscillating frequency, w, the amplitude of the potential applied to the end caps U , and the

Chupter 3

50

amplitude of the ring electrode voltage, VRF,according to equations 3.27 and 3.28. uz = -2a, = -1 6ze Ulrn(r,’

+ zo2)w2

(3.27)

ys = -2q, = 8ze V,$rn(r,’

+ zo2)w2

(3.28)

Again an ion stability diagram can be constructed to define the (a,q) coordinates for which ions are stored in the trap. Ions possessing values of a and q that give them both axial (along the z-axis between the caps) and radial (in the plane of the ring electrode) stability will remain trapped. In modern ion trap instruments, an inert gas such as helium (added to a pressure of about Torr, or 0.1 Pa) helps to store the ions by lowering their kinetic energy through collision with the gas. A unique feature of traps is that ions can be introduced and stored until a sufficient density is achieved for ejection and analysis. In conventional quadrupole and magnetic sector instruments, ions continuously pass through the mass spectrometer and their population is largely determined by the efficiency of the ionisation process. The ability to store ions, and the relatively short path to the detector that the ions follow in a quadrupole ion trap, can lead to very high detection sensitivities. Ion traps can also be exploited in studies of ion chemistry (see Chapter 6) by using longer containment times. In practice, it is necessary to balance the desire to store large numbers of ions in the trap with space charge eflects that arise from the repulsion of neighbouring ions. These repulsive forces cause ions to leave the trap and not be detected. It also perturbs the motion of ions in the trap resulting in degraded mass resolution and mass shifts that lead to large errors in mass measurements. Thus it is desirable to control the population of ions in the trap at all times during analysis. For this reason, automatic gain control (AGC) was developed to control the ion generation rate and couple it to the time period in which ions are introduced into the trap and stored. A typical operation of an ion trap involves lowering the end cap potential closest to the ion source to allow ions to enter the trap. This voltage is raised once sufficient numbers of ions are stored as determined by the AGC measurements. The ring electrode is held at an appropriate value to store ions over a range of mlz values. The potential on the second cap electrode is then lowered and the ring electrode voltage VRF is linearly ramped to eject ions of increasing mlz in turn onto the detector.

51

The Mass Spectrometer

3.3.5 Ion Cyclotron Resonance Traps that make use of a magnetic field are known as ion cyclotron resonance (ICR) mass analysers. The name derives from the frequency of an ion’s circular motion wc within a magnetic field, B. In these instruments the trap consists of a cubic, rectangular or cylindrical “box” with entrance and exit slits cut in opposite sides or plates (Figure 3.21).

I

t ransmitteroplates

I

l

o

receiver plates

Figure 3.21 Ion trap of an ion cyclotron resonance ( I C R ) mass spectrometer

In a magnetic field, B, an ion with a velocity v will adopt a circular trajectory with a radius r perpendicular to the field when the centripetal and centrifugal forces it experiences are equal (equation 3.29). zvB = (mv2)/r

(3.29)

The angular velocity of the ion perpendicular to the field is given by equation 3.30. wc = vlr

(3.30)

Substituting equation 3.30 into equation 3.29 and simplifying, equation 3.3 1 is derived. wc = zBlm

(3.31)

Thus an ion’s cyclotron frequency depends on its mass and charge but is independent of its velocity (equation 3.31) and the mlz ratio for an ion can be determined by measuring its cyclotron frequency. Ions of lower mlz have higher cyclotron frequencies (Figure 3.22A); ions with higher mlz have low cyclotron frequencies (Figure 3.22B). Note that oppositely charged ions would move with the same cyclotron frequency but in opposite directions were they both present in the trap.

52

Chapter 3

Figure 3.22 Ion cyclotrons of high,jrequemy or low mlz ( A ) and low frequeizcy or high m/z ( B )

Ions moving in a magnetic field adopt stable cyclotron orbits and can be stored (up to several hours!) provided the pressure is kept low Pa). They do not, however, generate any (typically (CH,),CH+ > CH,CH,+). The ability of hydrocarbons to undergo rearrangement reactions after ionisation is also well-known. Many large hydrocarbons fragment to ions corresponding to the formula C,H,+ (mlz 43) (Figure 5.1 ) associated with an isopropyl species that subsequently loses ethylene to form the methyl cation CH,+(equation 5.13).

Organic Muss Spectrometry

91 H+ >-

CH3'

+ CH2=CH2

(5.13)

Hydrogen rearrangements are rarely detected in saturated hydrocarbons but are observed in the EI mass spectra of unsaturated forms. A McLafferty-type rearrangement is often encountered for unsaturated hydrocarbons with the elimination of an alkene (equation 5.14). [H-CH2-CH2-CH2-CH=CHJf'

CH2=CH2+ [CH,=CH-CH,]+'

(5.14)

Despite these rearrangements, the fragmentations of straight-chain and branched aliphatic hydrocarbons can often be deciphered through a one-step dissociation pathway. The same is not true of cyclic hydrocarbons. Cyclohexane, for example, shows a loss of ethylene in its EI mass spectrum. Dissociation of a o-bond cleavage followed by homolytic cleavage can account for the formation of a distonic ion at mlz 56 (equation 5.15) that represents the base peak in the spectrum.

Unsaturated cyclic hydrocarbons also fragment through more complex dissociation pathways. Cyclohexene dissociates to yield major fragment ions at mlz 54 and 67. The latter is produced by the loss of a methyl radical after ring opening by o-bond cleavage and hydrogen atom migration (equation 5.16). The ion at mlz 54 is associated with the loss of ethylene and corresponds to the ionized form of 1,4-butadiene (equation 5.17). This decomposition is a retro-Diels Alder reaction.

(5.17 )

Chupter 5

92

Aromatic hydrocarbons are more resistant to fragmentation and their EI mass spectra are usually dominated by the molecular ion, as well as its multiply-charged forms. In the case of benzene, in addition to the molecular ion at mlz 78 that appears as the base peak in the spectrum, other ions appear at m/z 5 1 and 39. The latter ion is the doublycharged molecular ion C 6 H P . Beynon and Fontaine measured the energy lost during the dissociation of C 6 H P to CH,' and C,H,' as 2.8 eV consistent with the two charges in the parent being located over 5 A apart. This observation suggests the doubly-charged form of benzene has an open-configuration with one of the following forms (Figure 5.2).

Figure 5.2 Proposed structures f o r the ring-open configurations of' the doubly-charged ion of benzene

The fragment ion at mlz 51 can also be explained if dissociation of benzene is preceded by a ring opening reaction (equation 5.18). [CH,=CH-CH=CH-C=CH]+' CH,=CH'

-+ (5.18)

+ 'CH=CH-CrCH

Toluene dissociates by the loss of a hydrogen atom to yield a fragment ion at mlz 91. Deuterium-labelling experiments have shown that further decomposition of this ion occurs through the loss of ethylene (m/z 65) suggesting all seven hydrogen are equivalent. This supports the formation of a cyclic tropylium structure as an intermediate (equation 5.19). The proposed rearrangement mechanism for the formation of the cyclic tropylium ion is shown in equation 5.19.

L

m/z 91

m/z 91

m/z 65

The cyclic tropylium ion often appears in the spectra of aromatic hydrocarbons, but not all alkylbenzenes necessarily give rise to a substituted tropylium ion C,H,R+. Yet ring expansion processes have also

Orgunic Muss Spectrometry

93

been proposed in the fragmentation of heteroaromatic compounds. Alkylthiophenes, for example, are proposed to lose alkyl radicals through a thiapyrylium ion (equation 5.20).

(5.20)

5.3.2 Alcohols

Aliphatic alcohols frequently give rise to EI spectra in which the ion signal for the molecular ion is generally weak or not observed. This is consistent with their lower ionisation efficiencies compared with the corresponding alkane. The absence or diminished intensity of the molecular ion is also associated with the susceptibility of alcohols to fragmention. It is convenient to visualise many of these fragmentation reactions as proceeding from a molecular ion first formed by the loss of an electron from one of the lone electron pairs from the oxygen of the hydroxyl group. Aliphatic alcohols dissociate with the loss of a hydrogen atom, alkyl radical, water and alkenes such as ethylene. This is illustrated below where water loss proceeds through hydride ion migration to the hydroxyl group which neutralises the charge on oxygen.

(5.21)

The loss of a hyGroxyl radical, in contrast to water, 1s rarely observed. It is common, however, to detect ions corresponding to the loss of molecular hydrogen (HJ from the molecular ion. Unsaturated alcohols undergo similar processes. But-3-en- 1-01, and other long chain and branched unsaturated alcohols, also dissociate following a McLafferty rearrangment according to equation 5.22.

+*

HCECH

(5.22)

Chapter 5

94

The EI mass spectrum of cyclohexanol is dominated by a fragment ion at mlz 57 (the base peak). This ion is formed following ring opening hydrogen atom migration, and the subsequent loss of a methyl radical and ethylene.

(5.23)

Many EI mass spectra of phenols have also been recorded. Like other aromatics, the spectra are dominated by intense ion signals for the molecular ion. Phenols, however, show the unique loss of carbon monoxide (-28 u) that has been proposed to occur through a cyclohexadienone intermediate followed by a compacting of the ring size to release CO (equation 5.24).

(5.24)

This loss is typically accompanied by the loss of 29 mass units corresponding to the CHO unit. Other substituents on the aromatic ring can stabilise the ionic products and hence can drive the fragmentation process. Alkyl substituents positioned para to the phenolic hydroxyl group can lose fragments from the benzylic position since the resulting ion has a stabilized oxonium ion form (Figure 5.3). An equivalent ion cannot form if the alkyl substituent is located meta to the phenolic hydroxyl group.

+

Figure 5.3 Oxoniurn ion formed by the loss oj'alkyl substituents para to the phenolic hydroxyl group

Organic Muss Spectrometry

95

Shannon and others have studied in some detail the mass spectra of benzyl alcohols. These exhibit abundant ion signals associated with the molecular ion (mlz 108 in the case of benzyl alcohol) in addition to fragments formed by the loss of a hydrogen atom and carbon monoxide. This fragmentation has been proposed to occur through the formation of a seven-membered ring intermediate with the subsequent loss of CO to yield a benzenium ion (mlz 79) (equation 5.25). This ring can then lose molecular hydrogen to form C,H,+ (mlz 77). An additional ion in the mass spectrum of benzyl alcohol is due to the loss of hydroxyl radical to yield the benzyl cation (mlz 91) that can be stabilised through the formation of a tropylium ion discussed earlier. A weaker fragment corresponding to the successive loss of molecular hydrogen and a hydrogen atom from the side chain gives rise to the benzoyl cation C6H,CO+(mlz 105) (equation 5.26).

(5.26)

m/z 105

5.3.3 Ethers The molecular ion generally appears in greater abundance in the EI mass spectra of ethers relative to their corresponding alcohols. Like alcohols, however, a common cleavage observed for ethers involves fission of a a-bond. Asymmetrical ethers can give rise to two products by this process. The more substituted ions will tend to form preferentially as illustrated in equation 5.27 over 5.28. (CH3)2CH-O+'-CH,-CH3

++ (CH,),CH+ + '0-CHZCH,

+ (CH,),CH-0' + +CH,CH,

(5.27) (5.28)

Another fragmentation process observed for ethers is hydrogen atom migration to the charge site with loss of an alkene (equation 5.29).

Chapter 5

96

(CH3)2CH-O+'-CH,-CH,

3 CH,-CH=O+-CH,-CH,

+ CH3'

+ CH,-CH=OH+ + CH,=CH,

(5.29)

Unsaturated ethers, such as alkyl vinyl ethers, can undergo a McLafferty rearrangement resulting in the loss of an alkene and a stable carbonyl cation (equation 5.30). [H-CH,CH,-O-CH=CH,] CH,=CH,

+

+'

+ (5.30)

[O=CH-CH,I+'

Ethers possessing aromatic groups fragment through the simple a cleavage processes shown above. Phenyl ethers also have been observed to undergo secondary fragmentations involving the loss of carbon monoxide as illustrated in equation 5.3 1.

+'

0 & -co OCH3

0

m/z 93

8

(5.31)

m/z 65

This fragmentation is blocked when an additional methylene group is located between the oxygen atom and the aromatic ring. Such benzyl ethers have EI mass spectra that are dominated by the benzyl cation C,H,-CH,I ( m h91) formed by a-cleavage. 5.3.4 Amines

Aliphatic amines generally ionise poorly by EI. However, due to the basic nature of the amino groups, stable [M+H]+ions can be produced in high yield during chemical ionization. a-Cleavage is the predominate fragmentation pathway (shown for ethyl amine in equation 5.32) with P-cleavage, and to a lesser extent y-cleavage, becoming more important as the size of the carbon chain increases. [CH3CH2-NH2]+'+ CH3' + CH,=N+H2(mlz 30)

(5.32)

One of the most dominant rearrangment processes observed in the

97

Orgunic Muss Spectrometry

spectra of aliphatic amines, arises from the transfer of an alkyl group from the carbon a to the amine group to the equivalent carbon on the opposite side followed by cleavage of the N-C bond (equation 5.33). This fragmentation process is only possible for secondary and tertiary amines. [(CH,),CH-NH-CH,]+'

+ CH,CH=NH+ + 'CH2CHX

(5.33)

Aromatic amines, in contrast to aliphatic amines, give rise to dominant molecular ions in their EI spectra. In many cases, as in the spectrum of aniline, the molecular ion is the base peak of the spectrum. The formation and stability of the molecular ion is attributed to electron transfer with the x-electrons of the ring (Figure 5.4).

Figure 5.4 Proposed structures for the molecular ions of aniline that contribute to its stability by electron transfer with the aromatic ring

5.3.5 Aldehydes and Ketones

The addition of a carbonyl group to an alkane considerably lowers the ionisation energy of the molecule. The lowest energy form is that in which an electron is removed from one of the lone pairs of the carbonyl oxygen. Removal of an n-electron from the C=O bond requires more energy (typically 10.6 eV) than that required to remove an electron from a a-bond (1 1.5 eV). Since most EI spectra are recorded at 70 eV, all forms of the molecular ion are possible. A common dissociation pathway in both aldehydes and ketones is that which results from cleavage of the bond a to the carbonyl group. This results in the loss of a hydrogen atom in the case of aldehydes and alkyl groups in both systems with the formation of the resonance stabilized ion R-C+=O (equation 5.34). [R-CO-R']"

+ R-C+=O + H' or R"

(5.34)

The ion H-C+=O (mlz 29) is a signature fragment in the case of aldehydes and often appears as the base peak of the spectrum.

98

Chapter 5

A McLafferty rearrangement can arise in the case of long chain aldehydes, as illustrated for the butyraldehyde ion in equation 5.35. This results in the cleavage of the P-bond through a cyclic transition state and the loss of an alkene. The mass of the latter provides an indication of the degree of branching in the molecule.

The aromatic ketone benzophenone is dominated by a strong molecular ion (mlz 182). The base peak of the spectrum at mlz 105 arises from the benzoyl ion C,H,-C=O+ with a corresponding benzyl ion detected at mlz 77. The latter is generated as a secondary fragment of the benzoyl cation in the case of alkylphenones such as acetophenone.

5.3.6 Carboxylic Acids, Esters and Amides The spectra of aliphatic carboxylic acids and esters, like aldehydes and ketones, are dominated by ions associated with a McLafferty rearrangement and the loss of an alkene.

Both the resonance-stabilised O=C+-OH a +O=C-OH (mlz 45) and O=C+-OR ions predominate in the EI spectra of acids and esters respectively. Cleavage P to the carbonyl group gives rise to the resonancestabilised ions 'CH,C(=O)-OH and 'CH,-C(=O)-OR. The loss of water from a carboxylic acid usually requires an aliphatic chain of at least four carbon atoms long with hydrogen atom transfer from the y-position accompanied by the cleavage of C-H and C-0 bonds. The loss of a hydroxyl or alkoxide radical from a carboxylic acid or ester is favoured if the resulting product ion is stable. As an example, the EI mass spectrum of benzoic acid is dominated by fragments associated with the loss of the hydroxy radical and the subsequent loss of CO (equation 5.37). [C,H,-C(=O)-OH]'*

+ CsH,-C=O+ -+C6HSf

(5.37)

Organic Mass Spec trorne try

99

Deuterium-labelling studies, however, have shown that the loss of HO' to form the ion at mlz 105 does not solely involve the hydrogen atom of the carboxylic acid group. The EI spectrum of ortho-d-benzoic acid, for example, exhibits ions at both mlz 105 and 106 due to what is known as the ortho-eflect. The former product arises from transfer of a proximal hydrogen atom of the ring to the carboxylic acid group prior to hydroxyl radical loss (equation 5.38).

OD0 - 6 9

HO-C=O

HO-C-OD

-bD

&

\

OH

+

+*

/

\

m/z 105

(5.38)

As one would expect, the EI mass spectra of amides resemble that of their corresponding acid and ester. An additional process, in the case of secondary and tertiary amides, results from the cleavage of the N-C bond and the transfer of one or two hydrogen atoms to produce a neutral loss (R-C(=O)-NH,) and the ion R-C(=O)-NH,+.

5.3.7 Halides The halogen atoms F, C1, Br and I have a relatively small effect on the fragmentation processes of organic compounds. Molecular ions are detected in the EI mass spectra of halides though the proportion of charge residing at the halogen atom will vary (I>Br>Cl>F) counter to the electronegativityof the atoms. As a result, a-bond cleavage (equation 5.39) occurs preferentially adjacent to I. R-CH2-X"

+ R' + CH,=X'

(5.39)

The intensity of the X' ion also is observed to increase as the electronegativity of the atom decreases. Hence the I' ion appears in greater abundance than the F' ion. This is consistent with the formation of R' from the fragmentation of R-X where the loss of X' is observed to a greater extent in iodides and bromides, over chlorides and fluorides. This process is evident in the mass spectra of alkylhalides but not aromatic halides due to the stability of the C-X bond when the halogen is attached directly to the ring.

Chupter 5

100

The abundant isotopes of some halogen atoms provide a unique signature that aids in the determination of an unknown compound. All fragments containing one chlorine atom for instance should appear as two ions differing by two mass units in an approximate ratio of 3: 1 due to the natural abundance of 35Cland "Cl. The corresponding ions containing a bromine atom will appear two mass units apart in a ratio of 1: 1 due to the natural abundance of 79Brand *lBr(see Appendix 2). The loss of hydrogen halide is a common fragmentation pathway in the case of chlorides and fluorides (equation 5.40). R-CH,-CH,-X+*

+ [R-CH=CHJ" + HX

(5.40)

5.4 QUANTITATIVE ANALYSIS OF ORGANIC COMPOUNDS Having covered many of the common fragmentation processes observed for organic molecules with various functional groups, we turn our attention to the quantitative analysis of such compounds. Since organic compounds are used widely in prescription drugs and for agricultural, food and industrial purposes, the quantitation of such compounds in living systems, extracts and the environment is of particular importance. While the identification of an organic compound's structure provides valuable information, the absolute or relative levels of that compound in the sample may also be critical. For example, the identification of a performance-enhancing compound in the blood or urine of an athlete prior to competition may alone be sufficient to ban the athlete from competing. However, where the compound (such as a hormone) occurs naturally in the body it may be necessary to establish that a higher than usual dose has been administered. Mass spectrometry plays a major role in the quantitation of organic compounds in this, and many other, applications.

5.4.1 Role and Choice of Quantitation Standards Quantitative analysis of any compound by mass spectrometry first requires that the detector response be calibrated as a function of the concentration of a compound at a particular set of operating conditions (e.g. ionisation conditions, ion source settings, tuning parameters, etc.). This allows the ion current detected by the mass spectrometer to be reliably correlated with the amount of compound in a particular sample. Note that this may involve a reasonably large number of measurements, since the detector response does not necessarily vary in a linear manner as the sample concentration changes across several orders of magnitude.

Organic Muss Spectrometry

101

To perform the calibration optimally, the quantitation standards should have the same structural characteristics and be of a similar (though not the same) size to the compounds of interest. This ensures that the ionisation and detection efficiencies of the quantitation standards and the compounds under investigation are essentially identical. Ideally, the standards should be added to the sample; that is, they should be internal quantitation standards. This prevents any fluctuation in the performance of the instrument during the analysis of the standards, and subsequently the sample of interest, from adversely influencing the quantitation measurement. It is important that the internal standards be added to the sample at the earliest possible stage, so that they are subjected to the same potential losses prior to and during the analysis.

5.4.2 Calibration of the Detector Response This stage of the analysis involves measuring the ion current of a particular standard or set of standard compounds as a function of their concentration. Depending on the concentration variations predicted for the samples that are to be analysed, the concentration of the standards might be varied over one or several orders of magnitude. Where the measurements are performed on a mass spectrometer featuring a scanning mass analyser (magnetic or quadrupole-based, including ion traps), the instrument is operated in the selected ion monitoring (SIM) mode. In this mode, the mass analyser is scanned over very small rnlz ranges (about the ion signals of the standard(s)) to detect the majority of (if not all) the ions present. If a larger rnlz range were scanned, the ions produced from the standard compound(s) would be passed to the detector during only part of the scan; the time the ions are detected being dependent on the scan rate of the instrument. In other words, during most of the scan the mass analyser would attempt to transmit ions of different rnlz ratios than those of the standard@) to the detector. In the SIM mode, many scans and thus mass spectra are obtained within a particular time interval. Where ions from several quantitation standards are to be detected, the mass analyser is scanned segmentally over several small rnlz ranges about each ion. Once the detector responses are measured, the area under the ion signals can be plotted as a function of the concentration of the standard(s). Any deviation in the areas obtained in subsequent runs from a line-of-best-fit establishes the error of the analysis. It is desirable to use the area under, rather than the height of, an ion signal for quantitation because the latter is highly influenced by the mass resolution of the measurement. The height of an ion peak corresponding to a

102

Chupter 5

particular set of isotopes is less than that where the isotopes remain unresolved. It is typical in selecting the quantitation standards to choose compounds that have a range of molecular weights so as to produce ions across the mlz range of the instrument. This is because ion detectors do not detect all ions across the full mlz range of a particular mass analyser with equal efficiency. Low mlz ions are generally detected with much greater efficiency than high mlz ions, irrespective of whether they are produced in equal quantities in the ion source.

9000 8000 7000 6000 5000

4000 3000 2000 118

1000

'il

0 50

60 70 80

90 100 110 120 130 140 150 160 170 180

(B)

35000

Ion 98.00 30000

25000

I

Cotinine

Figure 5.5 EI mass spectrum of cotinine (Figure 5.5A) and selected ion chromatogram of fragment ion mlz 98 (Figure 5.5B)in the saliva of an active smoker (Source: J.-G. Kim, U.-S. Shin and H.-S. Shin, Rapid Monitoring Method of Active and Passive Smoker with Saliva Cotinine by Gas ChromatographyMass Spectrometry, Bull. Korean Chem. Soc., 2002, 23(10), p. 1497)

Organic Mass Spectrometry

103

5.4.3 Quantitative Analysis of Cotinine; Example of Selected Ion Monitoring Cotinine is a metabolite of nicotine that has been detected in both smokers and non-smokers by selected ion monitoring to assess the risk of passive exposure to cigarette smoke. The levels of cotinine were monitored in the blood, urine and saliva of both smokers and nonsmokers. The EI mass spectrum of cotinine exhibits a fragment ion rnlz 98 (the base peak) and a molecular ion at mlz 176 (Figure 5.5A). Selected ion monitoring of the base peak at mlz 98 thus affords optimal sensitivities for these experiments. A typical SIM ion chromatogram for the ion at rnlz 98 is shown in Figure 5.5B at a cotinine concentration of 128 ng ml-'. Cotinine levels were measured based upon the ratio of the fragment ion peak area of cotinine at mlz 98 relative to that of the internal standard d,-deuterocotinine (mlz 101) by interpolation from the regression line of the standard curve. Detection limits of 5-50 ng ml-' were achieved among the biological matrices. The precision of the quantitation measurements was reported to be between 83.9-99.8%.

FURTHER READING H. Budzikiewicz, C. Djerassi and D.W. Williams, Mass Spectrometry of Organic Compounds, John Wiley & Sons, New York, 1967. Q.N. Porter, Mass Spectrometry of Heterocyclic Compounds, John Wiley & Sons, New York, 1985. EW. McLafferty and F. Turecek, Interpretation of Mass Spectra, University Science Books, 1993.

CHAPTER 6

Ion Chemistry 6.1 ELECTRON AND PROTON AFFINITIES AND MEASUREMENTS OF GAS PHASE ACIDITY 6.1.1 Electron Affinity The most fundamental property of a molecule or atom from the perspective of mass spectrometry experiments is their ability to lose or gain electrons and form ions. The energy required for a molecule or atom to gain an electron is known as its electron ufinity (EA). This results in the formation of a radical-anion according to equation 2.2. The electron affinity of a molecule can be defined by equation 6.1. AH: is the heat of formation of an molecule or ion defined as the heat absorbed or released when one mole of the entity is formed at standard temperatures and pressures (298 K = 25 "C, 1 atmosphere = 1.013 x lo5 Pa). Values of AH: can be estimated by simple arithmetic using available thermodynamic data or by ub initio molecular orbital calculations. EA(M) = AH:( M) - AH:( M-.)

(6.1)

Two types of experiments are employed to determine electron affinities. The first of these uses photon detachment methods. Using a crossed photon-molecular beam apparatus, a photon source such as a laser intersects at right angles with a beam of negative ions. The production of neutral molecules according to equation 6.2 is then studied.

The neutral molecules produced are detected using a particle multiplier. Ion cyclotron resonance (ICR) mass spectrometers are also used for these measurements where the photon beam runs parallel to the ion motion. Due to the long trapping times, the spatial overlap of the ion 104

Ion Chemistry

105

and photon beams is relatively large and photon detachment efficiencies are high. The minimum photon energy required to remove an electron is measured in these experiments. Photon-ion interactions are particularly suited to measurements of electron affinities since the electron binding energies of most negative ions fall in the range of 0.5 to 3.0 eV, corresponding to approximately 400 to 2500 nm. The second type of experiment used involves a measurement of the energy required to transfer charge between a negative ion and a neutral molecule. In equation 6.3, this charge transfer is associated with the transfer of an electron from A-' to B.

Experiments to determine whether a series of ions will transfer an electron to a particular molecule enable the electron affinity of the molecule to be bracketed. As expected, molecules that contain electronegative atoms are more likely to bind electrons and thus will have higher electron affinities. The instability that results from pairing electrons in atomic orbitals, however, has an impact on this trend. The group XV atoms of the periodic table (nitrogen, phosphorous and arsenic) have much lower atomic electron affinities as a consequence of their half-filled p-orbitals. Stabilisation of the negative charge through conjugation can raise the electron affinity of a molecule. The phenoxide radical, for example, has a higher electron affinity (2.4 eV) than other alkoxide radicals due to the delocalisation of the charge on oxygen throughout the aromatic ring.

6.1.2 Gas Phase Acidity and Proton Affinity Measurements of electron affinities enable other thermodynamic data to be derived. For instance, the enthalpy contribution to the gas phase acidity of a compound can be derived from equation 6.4 where BDE is the bond dissociation energy of the M-H bond and IE is the ionisation energy for the molecule. The latter is defined as the energy required for the process defined by equation 2.1. AHaci2(M-H)= BDE(M-H) - EA(M) + IE(M)

(6.4)

Equation 6.4 is derived from a thermodynamic cycle produced from a sum of the following two processes (equations 6.5 and 6.6).

Chapter 6

106 M-H

+M + H

M + H --+ M-+ H'

= BDE(M-H)

(6.5)

AH: = E,(M) - EA(M)

(6.6)

AH:

Since the primary process by which most [M+H]' ions are formed during ionisation is by the transfer of a proton from one compound to another, a molecule's gas phase acidity is an important property in mass spectrometry. Proton transfer reactions are also among the most important processes in chemical and biochemical transformations and have been studied extensively in solution. The gas phase acidity of a neutral compound MH is defined as the free energy AGaCi2(MH)required to effect the forward reaction shown in equation 6.7. MH

M- + H'

(6.7)

The free energy for the reaction has both enthalpy and entropy contributions according to equation 6.8. AG,,,O(MH) = AH,,,O(MH)

-

TAS,,~:(MH)

(6.8)

The proton aflnity (PA) of a neutral molecule M is defined as the energy required to effect the forward reaction shown in equation 6.9. Thus PA(M) = AH,,;,O(MH+). -

The traditional approach to measure the gas phase acidity of the compound MH is to react it with a base (B) and measure the degree of formation of the ion BH' (equation 6.10). MH + B

M-+ BH'

(6.10)

When this measurement is performed in solution, the process is strongly influenced by the solvent medium. Proton transfer reactions with the solvent interfere with and limit the accuracy of such experiments. A mass spectrometer, by comparison, enables the intrinsic reactivity of a molecule to be studied in the absence of solvent and a number of specialised instruments for this purpose have been constructed. These instruments have led to the construction of tables of gas phase acidity data.

Ion Chemistry

107

6.1.3 Gas Phase Acidity Measurements Most measurements of gas phase acidities have been performed with high pressure mass spectrometers and both ion cyclotron resonance (ICR) and quadrupole ion trap (QIT) instruments. These latter instruments store ions for sufficient times to effect ion-molecule and ion-ion reactions. Reaction products and rates can be determined using these mass spectrometers. An example of the former type of mass spectrometer is a .flowingaftevglow instrument. These devices were originally constructed to study the chemistry of the ionosphere and feature a reaction flight tube along which reactive species can be added in a gaseous form. Their name derives from the visible glow detected within the earlier glass drift tubes from energy lost in exothermic reaction processes. The flight tube is pressurised with helium to collisionally 'cool' the ions formed in the ion source and ensure they are not in an excited (high energy) state. The ions produced in the source drift down the flight tube and react subject to the nature of the molecule added and the reaction time available. The length of the flight tube provides a reaction time domain along which rate constants can be derived. A common method used to determine the acidity of a gaseous molecule is through bracketing methods. If a molecule MH transfers a proton to base B, but not to base B,, its AGaCi2(MH)value will be greater than AGaci2(B1H)but less than AGaci2(B,H). In order to rank these acidity measurements, performed on many instruments at various temperatures, on a common AHacid'scale it is necessary to predict the entropy change ASaci: according to equation 6.1 1. ASaCi2(MH)= P(H')

+ P(M-) - P(MH)

(6.1 1)

For a known value of P(H'), the values of P(M-) and 5?(MH) are the most dissimilar in terms of their rotational contributions. According to theory, the entropy contributions from translational, vibrational and electronic effects can be considered to be identical for M- and MH and thus cancel each other out. The (usually small) entropy change for the reaction is then estimated from statistical mechanical considerations of rotational entropies. In practice, as these estimates can be unreliable and since most studies are conducted at a single temperature (25 "C), the small entropy contribution to a molecule's gas phase acidity is often ignored. A problem with bracketing experiments is that the errors in known acidity data are transposed into the measured data for unknowns. An alternate method to determine gas phase acidities is the kinetic method.

Chapter 6

108

6.1.4 Kinetic Method The application of the kinetic method to the measurement of gas phase acidities involves measuring the ratio of product ion signals from the competitive fragmentation of the dimeric precursor ion [A . . . H' . . . B]. This ion may undergo metastable decomposition or be activated to dissociate through collision-activated dissociation (CAD) or other process. Two product ions AH+and BH' are formed with rate constants of k , and kB(equation 6.12). [A . . . H' . . . B] + AH'

+ B with a rate constant kA

+ A + BH'

with a rate constant k,

(6.12)

The relative abundances of the product ions are related to their differences in acidity A(AHacid) (equation 6.13) for a proton dimer at temperature, T. R is the gas constant (8.3 x lo3J K-'). In (kA/k,)= In ([AH+]/[BH+]) =: A(AHacid)/RT

(6.13)

A calibration curve based on the ion ratios of several species with known acidities is used to establish measured values. The kinetic method assumes that the competitive dissociations of [ A . . . H' . . . B] have equal entropies and thus cancel each other out. The popularity of the kinetic method is the result of several features including a high degree of sensitivity to differences in structure (including isotopically-labelled forms), the close agreement in values obtained by the method with those from other approaches, and the speed with which the analyses can be performed. Since the measurements are performed on a tandem mass spectrometer, it is also not necessary that the samples be pure in order for them to be studied. 6.2 ION-MOLECULE REACTIONS 6.2.1 Types of Ion-Molecule Reactions There are many types of ion-molecule reactions including electron transfer (equation 6.3), proton transfer (equation 6.9), addition, substitution and elimination reactions. Addition reactions involve the formation of a new covalent bond, while substitution and elimination reactions are also characterised by bond cleavage. Nucleophilic substitution reactions are a class of ion-molecule reactions that have received much

Ion Chemistry

109

interest given that such processes are common in organic chemistry. In the gas-phase, in the absence of solvent, these reactions proceed along a reaction pathway with two low-energy intermediates shown in square brackets in equation 6.14. N- + M-X S [N- . . . M-XI S [N-M

. . . X-] + N-M + X-

(6.14)

It is convenient to present an ion-molecule reaction in terms of a potential energy diagram with the reaction co-ordinate on the x-axis. The curve (in two dimensions) or surface (in three dimensions) allows a reaction process or series of competing reaction pathways to be viewed in which the energy of the system never passes below the curve or surface. For the reaction shown in equation 6.14, a double energy minima potential energy profile is constructed (Figure 6.1) featuring a central energy barrier that represents the transition state [N . . . M . . . XI-*. This transition state may reside at an energy that is above or below that of the reactants.

t

N- + M-X

energy

[N-M...X-] reaction co-ordinate

+

Figure 6.1 Energy projile for an exothermic ion-molecule reaction without an activation energy barrier

Gas-phase ion-molecule reactions importantly enable the role of solvent molecules in a chemical transformation to be explored. In the presence of solvent vapour in the mass spectrometer, the reaction of an ion with increasing levels of solvation M'(S), or M-(S), (where S is a molecule of solvent) can be explored. Historically, most ion-molecule reactions have involved the reaction of singly-charged ions (either positively or negatively charged) and a neutral

110

Chapter 6

molecule though an increasing focus concerns the reaction of multiplycharged ions given their ease of production by electrospray ionisation. The reaction of multiply-charged ions with neutral molecules provides information about the site of the charges within the ion and whether such charged centres are localised or delocalised.

6.2.2 Rates of Ion-Molecule Reactions Ion-molecule reactions are one of the fastest chemical reactions known. This is a result of attraction between the charge of the ion and the dipole of a polar molecule. This electrostatic interaction is sufficient to overcome many energy barriers to reaction products. As a result, many ion-molecule reactions are exothermic and rapidly proceed even at room temperatures. An exothermic ion-molecule reaction that proceeds without an activation energy barrier has a rate constant k that is equal to the collision-control or diffusion rate (Figure 6.1). That is, the reaction is limited only by the ability of an ion and molecule to encounter one another. As these species do so, an electrostatically-induced or natural ion-dipole interaction occurs resulting in the formation of the ion neutral complex and subsequently products. Collision-controlled rate constants are of the order of I 0-9 cm3molecule-' s-'.

6.2.3 Ion-Neutral Intermediate Complexes It is common when analysing most ion-molecule reactions in the gas phase to infer the formation of an intermediate complex. Zon-neutral complexes have been postulated for many reactions in an attempt to explain the mechanism of formation of unusual product ions observed within mass spectrometers. Electrostatic considerations demand that when a gas-phase ion encounters a neutral species, its energy initially decreases. Thus an ionneutral complex is formed before any energy barrier is encountered in order to generate products. In the classical S,2 nucleophilic substitution reaction where X- reacts with CH3Y, the ion-neutral complex formed is represented by [X-CH,Y]- (see Figure 6.1). This results in a transition state structure denoted [X-CH,-.Y]- and the formation of an ionneutral complex for the product [XCH,-.-Y]-.This complex then dissociates to form the products XCH, and Y-. Square brackets are used throughout. The two components of an ion-neutral complex are associated not by a covalent bond but an ion-dipole attraction. Ion-neutral complexes can be formed from positively or negatively charged ions. The reactants are able to rotate about one another resulting in reactions that would

Ion Chemistry

111

not be geometrically possible if the partners were covalently bonded. In the gas-phase these species can be relatively long-lived with lifetimes of typically 10- 100 ps. Ion-neutral complexes are particularly significant for neutral species with a small dipole moment but a large polarisability. They typically have stabilisation energies in the range of 50 kJ mol-’; that is, some additional 50 kJ mol-’ is required for the complex to proceed to a transition state and ultimately to products. Ion-neutral complexes lose their relevance when solvation is possible. In contrast to the reaction profile above, the S,2 reaction between X- and CH,Y in solution proceeds simply from reactants to the transition structure [X.-CH3--Y]- and then to products. Ion-neutral complexes have been postulated as intermediates in elaborate mechanisms that attempt to explain the formation of unusual products in gas-phase ionic reactions and rearrangements. Molecular orbital calculations have also been used to implicate ion-neutral complexes in the fragmentation of certain ions. However, since these intermediates cannot be observed directly by spectroscopic methods, their existence is largely inferred rather than strictly proven. Ion-ion and ion-radical intermediate complexes have also been inferred in many gas-phase ion reactions. 6.3 KINETIC ISOTOPE EFFECTS When the substitution of an atom in a molecule or ion by its isotope alters the reaction rate of that molecule or ion, a kinetic isotope eflect (KIE) exists. The kinetic isotope effect is measured as the ratio of the rate constants for these reactions at a given internal energy. Where the isotopically-substituted atom is directly involved in bond dissociation or formation, the kinetic isotope effect is a primary one. When the atom is remote from the reaction centre, a secondary kinetic isotope effect is observed. The value of the kinetic isotope effect provides important information about the particular atoms participating in a reaction, distinguishing hydrogen exchange or scrambling, identifying a stepwise reaction mechanism from a concerted one-step process, and postulating the structure for a transition state. Most kinetic isotope effects are measured for the substitution of hydrogen with deuterium, i.e. k,lk,. The value of KIE increases from inverse (k,lk, < 1) toward normal (k,/k, > 1) as the “looseness” of the transition state increases. For reactions involving the reversible exchange of isotopes between molecular species, the kinetic isotope effect is given by the equilibrium reaction constant (equation 6.15).

Chapter 6

112

MH'

+ A-D

w MD'

+ A-H

K = k,lk,

(6.15)

An intermolecular isotope effect exists where an isotopically-labelled ion has two possible dissociation pathways (equation 6.16). D-CH,CH,Cl+'

+ CDH=CH,+'+ HC1

(6.16)

+ CH2=CH," + DCI Since this reaction occurs from a common precursor, the size of the isotope effect can be predicted from the relative abundances of the product ions.

FURTHER READING M.T. Bowers (ed) Gas Phase Ion Chemistry, Academic Press, New York, 1979. J.H. Futrell (ed) Gas Phase Ion Chemistry and Mass Spectrometry, John Wiley & Sons, New York, 1986. R.D. Bowen, Ion-Neutral Complexes, Acc. Chem. R e x , 1991, 24, 364-371. R.G. Cooks, J.S. Patrick, T. Kotiaho and S.A. McLuckey, Thermochemical determinations by the kinetic method. Mass Spectrom. Rev., 1994, 13(4), 287-339.

CHAPTER 7

Biological Mass Spectrometry 7.1 IONISATION OF BIOMOLECULES AND BIOPOLYMERS The low volatility and polar character of biomolecules and biopolymers initially prevented their direct ionisation and analysis by mass spectrometry. These compounds could, at best, only be studied after derivatisation of their polar groups, through methylation and acetylation, or following their degradation by, for example, acid hydrolysis. This converts large biopolymers into manageable (and ionisable) smaller molecules. Even then, only low to moderate (- 1000) molecular weight compounds could be introduced into a mass spectrometer in the form of gaseous ions. This situation changed, first with the development of plasma desorption and fast atom bombardment, and subsequently with the introduction of the electrospray and matrix-assisted laser desorption ionisation techniques. ESI and MALDI are particularly proficient at ionising large biopolymers (to several hundred thousand Daltons) without any pre-treatment or degradation of the sample. These ionisation methods are highly complementary in terms of their performance and suitability to particular samples. As a consequence, most laboratories that study biological compounds by mass spectrometry possess at least two instruments, one with an ESI source and the other with a MALDI source. Alternatively an instrument that can support both ion sources interchangeably can be used. Peptides, proteins, glycoproteins, glycoconjugates, glycolipids, lipids, oligonucleotides and moderately-sized nucleic acids can all be efficiently introduced into a mass spectrometer by virtue of the ESI and MALDI techniques. Their importance to the analysis of biological macromolecules was recognised in 2002 by the award of the Nobel Prize in Chemistry to those who contributed to their discovery, John Fenn and Koichi Tanaka.

113

114

Chup ter 7

7.2 PEPTIDES AND PROTEINS 7.2.1 Molecular Weight Analysis Following Barber’s demonstration that peptides and small proteins could be successfully ionized by FAB in 1980, the field of protein mass spectrometry has rapidly developed. Mass spectrometry is often the first approach employed to characterise protein samples and now can provide a great deal of structural information. This includes measuring the size of a protein, its complete amino acid sequence, the nature and site of post-translational modifications and even three-dimensional structural characteristics of a protein. Recent work has led to the use of mass spectrometry for studying protein interactions. Molecular weights are routinely measured to an accuracy of better than 0.01%, or 1 Da at a molecular weight of 10 kDa. The use of ESI and MALDI coupled to high resolution mass spectrometers, such as FT-ICR instruments (Chapter 3), has enabled the molecular weights of proteins (and other biopolymers) to be measured to an accuracy of a few ppm. This is illustrated in Figure 7.1 for the protein chondroitinase of 112 kDa measured to an accuracy of 3 Da (27 ppm). The mass accuracies obtained in all cases are far superior to those from other methods, including gel electrophoresis.

900

1100

.

1300

1500

.

Figure 7.1 The ESI FT-ICR muss spectrum ofthe protein chondroitinase of moleculur weight 112 kDu measured to un uccuvucy of 1 Da (ov 9ppm) (Source: McLafferty et al., 1 Am, Soc. Muss Spectrom., 1997,8, 380-383, Figure 1 - Part A)

Molecular weight measurements provide useful information in their own right, and may indicate that the protein isolated from a biological sample is different or in a modified form from that expected. Heterogeneous proteins and protein mixtures are often encountered in mass spectral data even when an investigator might believe the sample to be

Biological Mass Spectrometry

115

pure. Mass spectrometry can also analyse a complex mixture of many different proteins in a single analysis and with high sample throughput, both features that are important to proteomic discoveries discussed later in this chapter. A number of web-based algorithms have been developed to search protein databases with molecular weight information. Yet these databases contain a large number of proteins that coincidentally or due to structural similarities can share very similar, or even identical, molecular weights. As such it is often not possible to use the molecular weight of a protein alone to unequivocally identify it. For this reason, databases can be searched using a range of available information about a protein including its biological source, p l , molecular weight or features of its structure.

7.2.2 Mass Mapping To assist with the identification of a protein that may already be known or display similar sequence homology to a protein that appears in a database, apeptide muss map can be of use. Here the protein is digested with a site-specific protease such as trypsin (which cleaves proteins on the C-terminal side of arginine and lysine residues) and the peptide products are analysed collectively by mass spectrometry (Figure 7.2). The molecular weights of these proteolytic peptides, in addition to that for the intact protein, are searched against the mass of the theoretical peptide products generated by the same enzyme for all proteins in the database. The protein@) with the closest match based on the number and closeness of the masses matched appears in the output of the algorithm with the highest score. A few important points must be made in relation to these searches. First, a match in the mass of a peptide with a hypothetical fragment of a known protein does not alone prove that the peptide has an identical sequence. Peptides identical in mass may coincidentally possess the same mass even when they have quite different sequences. Furthermore, some amino acid residues are indistinguishable by mass (see Appendix 8 for glutamic acid and lysine, leucine and isoleucine) and could be interchanged within a protein with no change to its molecular weight. A second important issue concerns the level of protein coverage represented by the map. Ideally, the mass spectrum should contain ion signals for all peptides across the entire sequence of the protein. In practice, due to ionisation and detection efficiencies and the ease with which some sites are cleaved by enzymes over others, the map reflects only part of the total protein. Coverage levels vary depending on the complexity of the sample,

116

Chupter 7

4

p3

:.

0

4 2 a

2 1

lo00

1500

2000

2500

Figure 7.2 IdentiJication of yeast protein ILV5 jiom a MALDI mass map by database searching. Ions whose measured tnlz are M9ithin 50 pprn of’ those culculated are indicated by circles. (Source: A. Shevchenko, O.N. Jensen, A.V. Podtelejnikov, F. Sagliocco, M. Wilm 0.Vorm, P. Mortensen, A. Shevchenko, H. Boucherie and M. Mann, Linking genome and proteome by mass spectrometry: large-scale identification of yeast proteins from two dimensional gels, Proc. Natl. Acad. Sci. USA, 1996,93(25), 14440-14445)

the nature of the proteins and the type of mass spectrometer. Thus a match of masses for peptides that only represent a portion of the total protein with those theoretically generated for a database entry leads to the possibility that they differ within other regions. A third issue to consider when identifying proteins is mass accuracy. Irrespective of the type of mass spectrometer and measurement, some mass error is inevitable and the assignment of a protein to that of a database entry inherits these errors. This last consideration is becoming less important given that the molecular weights of peptides can be routinely measured with high accuracies. Mass measurements accurate to a few ppm are not uncommon and can be obtained even on time-offlight instruments employing ion mirrors and time-lag focusing where appropriate mass calibration procedures are employed. A number of web-based algorithms that use peptide mass maps to search protein databases are publicly available. These include Mascot at Matrix Science and Mass Mapper in the UK, Peptide Search at the European Molecular Biology Laboratory (EMBL), and Protein Prospector and PROWL developed in the United States. In many cases, mirror sites have been established at other laboratories to assist with data transfer throughout the world. Uniform resource locator (URL) addresses for these sites appear in Appendix 1 1. An example of the output from the ProFound algorithm at the PROWL web site based on a search of the Swiss-Prot database for some

Biological Mass Spectrometry

117

of the peptide mass map data (mlz 1547.8, 1612.7 and 2145.1) from Figure 7.2 is shown in Figure 7.3. A mass error of 0.05 Da was selected and the cysteine residues were deemed to be unmodified. A protein molecular weight range of 20-70 kDa was chosen but no p l range selected.

Figure 7.3 The ten highest-ranking entries output by the PROFOUND algorithm from a seurch ofthe S WISS-PROT database with three mlz values (mlz 1,547.8, 1,612.7 and 2,145.1) of tryptic peptide ions generated from a yeast protein IL v5

The yeast protein IVL5 (entry P06168) appears as the fifth entry in the output. The use of a larger set of mlz values of ions from the mass map allows this protein to be identified with a greater confidence. 7.2.3 Peptide and Protein Sequencing

A protein can only be unequivocally identified when its entire amino acid sequence has been determined. Peptide and protein sequencing can now be accomplished solely within the confines of a mass spectrometer, or by using chemical and enzymatic approaches in conjunction with mass spectrometric analysis. 7.2.3.1 Chemical and Enzymatic Sequencing In these approaches, the peptide or protein under investigation must be in a purified form. Other contaminating compounds in the sample can seriously compromise the

118

Chapter 7

analysis. The peptide or protein is treated with a chemical or enzyme to cleave amino acid residues from the N or C-terminus in a stepwise manner. An aliquot of the sample is removed at a series of time points and the collections combined. A mass spectrum is then recorded of the combined reaction products. The difference in mass of the products should correspond to the residue masses of the amino acids (Appendix 8) representing the molecular weight of an amino acid less 18 u for a molecule of water released in each step. This procedure has been described as ladder sequencing with the sequence of a peptide or the protein termini read directly from the mass spectrum. A partial C-terminal sequence (WCND) for an epitopic peptide of hen lysozyme has been determined based on its partial digestion with carboxypeptidase Y after reduction and alkylation of its cystein residues with ethyl pyridine (C*) (Figure 7.4). 14000

*

YG IFQI NSRYWCND

12000

*

'

10000

-

.n

I

(-WCND) j(

(-CND)

'

1

9

2

I

8000

(-D)

-2a

c

0

L> 0

600C

JOO[

2000

-_ 1do0

do0

1do0

Id00

Mass (mh)

2dOO

220D

Figure 7.4 MALDI mass spectrum of' the products of limited proteolysis oj-an epitopic peptide of hen lysozyme with carboxypeptidase Y (Source: J.G. Kiselar and K.M. Downard, Anal. Chem., 1999,71, 1792-1801, Figure 2)

Chemical methods employ the first Edman degradation reaction in which the N-terminus of the protein is converted to a phenylisothiocyanate (PTTC) derivate. The modified N-terminal amino acid residue is then cleaved with trifluoroacetic acid and the process repeated. This approach is not suitable for the sequencing of N-terminally blocked proteins or peptides as the initial PITC derivative cannot be formed. The enzymatic approach alternatively makes use of amino or carboxypeptidases to cleave amino acids from the N and C-termini respectively. In practice, the efficiency of a chemical or enzymatic cleavage

Biological Mass Spectrometry

119

decreases with the length of the polypeptide chain. Therefore, such an approach is capable of generating a complete amino acid sequence of a peptide, but not for a protein. To obtain the complete sequence for a protein, tandem mass spectrometry can be employed. 7.2.3.2 Tandem Mass Spectrometric Sequencing The sequencing of proteins by tandem mass spectrometry follows many years of study into the dissociation pathways of peptides under collisional and other ion activation conditions. Peptides have been shown to fragment along predictable pathways, for the most part involving the cleavage of the peptide backbone. The common fragment ions for a dissociated peptide are summarised in Figure 7.5 where a nomenclature has been adopted to identify those fragments that contain the charge in the N-terminal portion (a,, b,, c,, d,) or in the C-terminal portion (xn, y,, z,, v,, w,) of the peptide. Lower case letters are preferred so that they are not confused with nomenclature used for amino acids in their single-letter code. Many fragmentation processes involve the additional transfer of hydrogen atoms and protons but these are usually ignored for the purposes of labelling peaks in the MS/MS spectra of peptides to minimise the level of annotation. The numeral subscript (n) denotes the number of amino acid residues from either the N or C-terminus to the cleavage site. For example, a b, ion is formed by cleavage of the amide N-C bond at the 4th residue from the N-terminus. The fragment ions (dn, v,, w,) that are formed from the cleavage of both the backbone and a sidechain group are produced only at high collision energies (keV) where precursor ions are accelerated from the ion source at kV potentials. These fragments are only observed in tandem experiments performed on magnetic sector and time-of-flightbased instruments. Although at first sight they appear to complicate a tandem MS/MS spectrum, these side-chain specific fragment ions are useful to distinguish between isobaric residues such as leucine and isoleucine that exhibit unique side chain losses. Ideally a tandem MS/MS spectrum will contain only one series of ions in which the mlz value differences between each successive fragment corresponds to the mass of an amino acid residue (Appendix 8). In practice, a number of different fragment ion types can be produced simultaneously from discrete precursor ions due to the distinct structure of amino acids and the nature and energetics of the dissociation event. As certain bonds are more easily broken, and some product ions are more stable than others, the energy transferred and distributed throughout a peptide ion during activation influences its fragmentation. These factors are most pronounced in high energy (keV) dissociation experiments

120

Chapter 7

R

I

CHR’ v -NH

+2H

Figure 1.5 Common fragment ions jormed upon the dissociation ofpeptide or protein ions. Fragment ions containing the N-terminal portion of the peptide or protein are denoted a, h, c or d, those containing the C-terminal portion are denoted x,y, z, v or iv. The d, v and w ions are only.formed in high (keV) energy dissociution experiinen ts

where the nature of the fragments formed is strongly driven by the location of basic amino acid residues (particularly arginine and lysine). Low energy (eV) dissociation experiments, in contrast, generally give rise to mostly b,, and yn type fragments regardless of the peptide sequence (see Figure 7.6). Proteins can be sequenced following their treatment with a site-specific protease. MS/MS spectra are then acquired for each of the proteolytic peptides without the need for their purification. This establishes the sequence of segments of the protein but does not determine the order in which the segments appear in the molecule. A second protease of different specificity is used to generate a complementary set of peptides whose mass-to-charge ratios alone or in conjunction with their sequences derived from earlier MS/MS experiments allows the entire protein sequence to be assembled. Recent advances have overcome the need to digest a protein in order to obtain sequence data. Using a FT-ICR mass spectrometer, McLafferty and colleagues have performed tandem MS/MS experiments on the multiply-charged ions of intact proteins using electron-capture dissociation. This so-called “top-down” approach has led to the production of fragment ions that cover almost all of the protein’s sequence (see Figure 4.3). The great challenge rests with interpreting the single MWMS spectrum of the protein ions to derive the sequence. Although many of the fragment ions support more than one charge, the values for these can be measured based on the difference in the mass-to-charge ratios within their isotope distributions which are easily resolved on the FT-ICR

Biological Muss Spectrometry

121

instrument. However, since many of the fragments represent a large segment of the protein, the probability that their ions contain no atoms becomes exceedingly small. Thus many of the isotope distributions contain little to no detectable levels of ions containing no I3C (the monoisotopic or "C-only ion peaks). It is therefore necessary to measure the mass of the fragment based on a I3C-containing ion peak. The number of 13Catoms in the ion must be known in order for this mass measurement to be reliable. This has been approached by comparing the ion intensities within the resolved isotopic distributions with those generated theoretically. The closest match between an experimental and theoretically calculated profile allows the 13C in the ion to be assigned and the mass of the fragment to thus be derived. These top-down experiments are an impressive demonstration of the performance of an FT-ICR mass spectrometer. The use of an electrospray ion source provides an efficient means to introduce proteins directly into the mass spectrometer. The high resolution and ion storage capabilities of the instrument coupled with electron-capture dissociation allow the protein ions to be efficiently dissociated and the fragments reso1ved . 7.2.3.3 Interpretation of MSIMS Spectra of Peptides Although the methods described in this section have been developed to interpret the tandem (MS/MS) mass spectra of peptides, they can be extended to sequence proteins. Both manual and computer-assisted approaches are now in use to interpret the tandem mass spectra of peptides. The assisted approaches range from algorithms that can generate probable peptide sequences from the MS/MS data to those that attempt to identify the peptide, and the protein from which it may have been derived, by comparing the MS/MS spectral profile with a set of hypothetically-generated MS/MS spectra for all proteolytic peptides with the same mass across proteins of a database. Computer-based methods have considerably aided in the interpretation of MS/MS spectra, but it is important to note that they are fdlible and programs are known to assign incorrect sequences on some occasions. It is useful then to be practiced in interpreting MS/MS spectra of peptides by manual means. One approach to do so is illustrated for the data shown in Figure 7.6 recorded under low energy collision conditions. All mass-to-charge ratios represent monoisotopic (I2Conly) values. The MS/MS spectrum exhibits a series of fragments from m/z 175 to 1,51 1 in addition to a doubly-protonated precursor ion at m/z 813.2. The monoisotopic mass of the peptide is then 1,624.4 (or 813.2 x 2 (to correct for the charge z = 2) - 2 (for the mass of the protons attached)). Note also

122

Chuprer 7 63000

56000

79

-

4

~

8

5

49000. 42000

-

g

-

P 3 .:

35000-

22[.1

447.2 I

1122.8

I

584.3 1008.5

28000.

1251.6

2t000-

14000

1390.7

-

7000.

i 200

400

600

800

1

1000

1200

1400

Figure 7.6 Low energy CID tandem MSIMS spectrum of the doubly-protonated precursor ( P ) ions (mlz 813.2) ofpeptide X F E N X T P X H A N S R . X denotes either leucine or isoleucine (adapted from J.R. Chapman, ed., Peptide and Protein Analysis bey Mass Spectrometry, Ch. 6, Fig. 6, Humana Press, NJ, 1996, p. 95)

that all of the fragment ions with rnlz > 813 must be singly-charged since their mass is less than that of the intact peptide. A flow chart (Figure 7.7) can be constructed where amino acid residue mass (Appendix 8) differences are searched for between one fragment ion and the next, beginning with the one with the highest mlz value. The lack of a mass match ends a branch, and another possible association must be considered. Starting at the fragment ion with the highest mlz of 1,511.8, the subtraction of 113.1 (corresponding to either leucine or isoleucine) arrives at rnlz 1,398.7. Subtracting a further 147.1 units (consistent with the residue mass of phenylalanine), leads to a product ion at rnlz of 1,251.6. Repeating the process further, and ignoring the precursor ion signal, the sequence of amino acids can be read in single letter code as XFENXTP where X = I or L. Two ions appear below the ion at rnlz 697.4 with rnlz values of 617.3 and 584.3. The first of these corresponds to a mass loss of 81 u that is inconsistent with an amino acid residue. The second corresponds to a mass loss of 113 u consistent with leucine or isoleucine. Thus we derive the partial sequence XFENXTPX. It is now necessary to consider the fragment ions of low mlz. By repeating the subtractive process further, the sequence can be extended to XFENXTPXHANS down to the ion at rnlz 175.1. This remaining mass is associated with at least one amino acid residue. If the ion at mlz 175 is a b, ion, the N-terminal residue must have a mass of (175 - 1) (for the N-terminal hydrogen atom) or 174. Since this value is not consistent with the mass of an amino acid residue (Appendix 8), the ion

Biological Muss Spectrometry

123

h113.1 A147.1 A129.2 1511.8 + 1398.7 + 1251.6 + 1122.6

A80.9 617.3

A97.0 697.4

t--

794.4

lo@ 1

503.3

A113.1 Xle A81 .O 584.3

391.2

A137.0 His A56.0 'A70.9 447.2 + 376.3

@

@

AIOl.l

A1 13.0

@/All4.1

+ 895.5

f--

A1 14.2

A87.0

+ 262.1

+

1008.5

175.1

227.1 Figure 7.7 Interpretation of the low energy CID tandem MSIMS spectrum shown in Figure 7.6

must be considered to be a y, ion. The C-terminal residue must then have a mass of ((175 - 17) (for the mass of the C-terminal HO group) 2) (for the mass of the two protons attached to the ion) or 156. This mass is consistent with that of arginine. For the interpretation to be correct, all ions must be of the y,-series and thus the direction of the sequence from the N to C-terminus is XFENXTPXHANSR. It is not possible to distinguish leucine and isoleucine residues from MS/MS spectra recorded under low-energy conditions. These residues can only be distinguished in high-energy experiments where side chain (dn, w,) fragments are produced. As illustrated above, the interpretation of tandem mass spectral data for peptides involves locating fragment ions that are separated by approximately 100 u (or strictly 57 to 186 u) consistent with the residue masses of the amino acids. Mass loss differences greater than 186 may indicate that the ions are not from the same series or type, or that one or several ions of a fragment ion series do not appear in the spectrum. For example, a spectrum may exhibit a b, and b6 ion, but not a b, ion associated with cleavage of the 5th residue from the N-terminus at the amide bond. This b ion may not form due to energetic or stability issues. Under these circumstances, it is sometimes still possible to theorise as to the identity of the missing sequence based on the mass of dipeptides. It is not

124

Chapter 7

possible, however, without additional information to assign the order of these two amino acid residues. 7.2.3.4 Detection of’ Mutants and Post- Translational Mod8cations by MSlMS It is immediately apparent that MS/MS spectra can also be used to detect the presence of amino acid substitutions within homologous proteins (mutants) and post-translational modifications of amino acid side chains. Any such alterations will lead to a change in the mlz of all fragment ions that contain these amino acid residues. Fragment ions that do not contain the substituted residue or modification will appear at the same mlz ratio as that for the unmodified proteolytic peptide. However, post-translational modifications are often incomplete such that peptides that contain these modified residues are present at low levels in a protein digest versus their unmodified counterparts. Furthermore, some post-translational modifications (such as phosphorylation) can adversely impact the ionisation efficiency of peptides. A number of tandem mass spectral approaches have been implemented to assist with the detection and analysis of these modified peptides and their protein counterparts. Where phosphorylated peptides are ionised and detected in the negative ion mode, a series of characteristic ions, H,PO,- (mlz 97), PO< (mlz 79) and PO; (mlz 63) at low mlz ratios are detected. Alternately, a precursor ion scan (Section 4.3.2.2) to identify all peptides that dissociate to form these fragment ions can be employed. A neutral loss scan (Section 4.3.2.3) can also be of use to detect phosphopeptides due to the characteristic loss of 98 u associated with H,PO,.

7.2.4 Protein Structure and Folding Beyond the size and sequence of a protein, mass spectrometry can provide insights into a protein’s secondary and tertiary structure. The ability to detect differences in a protein’s conformational state came to light following the development of the ESI technique. Early studies of proteins showed that the charge state distribution of ions for a protein in its native state was centred at a higher mlz than that for the same protein in a denatured state. This was shown both as a function of pH, and more convincingly following reduction of a protein’s disulphide bonds (Figure 7.8). The relationship between a protein’s conformational state in the gas phase (in the absence of solvent) and that in solution is the subject of current investigation and debate. To probe the solution state of a protein, a number of indirect approaches that utilise mass spectrometry have been developed. Foremost among these is the use of hydrogeddeuterium exchange.

Biological Muss Spec t ronw try

125

Hen Egg W h i t e L y s o z y m e

-

1103

A

I M W 14,3061 12+

1

B 15+

1 'T , I , ,,

iI:,>\

,,

, ,

7y 7m 717

00

10'

20+

800

1000

1200

1400

1600

1800

Figure 7.8 ESI Y ~ Z U Sspectra S of hen egg white lysozynw in 5% acetic acid ( A ) and following the addition of ditlziothreitol ( B ) (Source: J.A. Loo, C.E. Edmonds, H.R. Udseth and R.D. Smith, Anal. Chem., 1990,62, 693-698, Figure 2)

7.2.4.1 Hydrogen Exchange Mass Spectrometry In these experiments, some of the hydrogen atoms in proteins are exchanged with deuterium or tritium by dissolving proteins in an isotopically-enriched solvent (such as deuterium oxide, D,O). The exchange of hydrogen occurs at different rates that are dependent in part upon the accessibility of each atom to the bulk solvent. Hydrogen atoms covalently bonded to carbon undergo isotopic exchange so slowly that this exchange is not observed. Hydrogen atoms of the hydroxyl, sulphydryl, amine and carboxylic acid groups of amino acid side chains exchange very rapidly (less than a second) and their rates of exchange are indistinguishable. Hydrogen of the amide backbone, however, undergo isotopic exchange at different rates ( k - 5 to 0.05 s-l) that can be measured and compared. Since each amino acid, with the exception of proline, has one amide-hydrogen located in the protein backbone, hydrogen exchange levels and rates can be measured along the entire length of the protein molecule. As these backbone amide hydrogen atoms participate in formation of secondary structural elements (such as alpha helices and beta sheets), the exchange rates are affected markedly by the structure and stability of the protein. At pH 7,

126

Chapter 7

the half-lives for isotopic exchange of amide hydrogens within a protein may be as short as seconds or as long as several months. The basis of the experiment is outlined in Figure 7.9. First a protein in solution at physiological pH (or some other pH of interest) is allowed to undergo hydrogen atom exchange with deuterium oxide. The reaction is then quenched by lowering the pH to 2.5 by the addition of acid and lowering the temperature to 0 "C. Under these conditions, the fast-exchanging hydrogen of the amino acid side chains are left unlabelled, while the slow-exchanging amide hydrogen atoms remain either substituted with deuterium or unexchanged. The entire protein is then analysed by mass spectrometry where the total level of deuterium incorporated is measured based on the increase in its molecular weight. For each deuterium atom that replaces hydrogen, the molecular weight of the protein increases by 1 Da. To determine the site of deuterium incorporation, a portion of the quenched sample is digested with the non-specific protease pepsin. This protease cleaves the protein efficiently at low temperature at residues across the entire protein sequence. The peptide segments are then analysed by mass spectrometry and their molecular weights measured. The level of deuterium incorporated into each peptide segment provides a way in which to measure the accessibility of the amide hydrogen to the bulk solvent across the entire protein in its original structure. Those peptides containing more deuterium represent a region of the protein backbone that is more accessible to solvent. Those peptide segments that contain minimal to no deuterium are interpreted as buried or shielded from solvent within the protein structure. In this way, the structure of a protein can be explored by mass spectrometry under a variety of solution conditions. By extension, this approach has been used to study the interaction of proteins with other molecules. Amide hydrogen in regions of the protein within the binding site will undergo isotopic exchange at a slower rate when shielded from the bulk solvent by the interacting molecule. When such an association does not occur, the same region will exchange hydrogen with the solvent more rapidly. Thus the location of the binding site can be determined. One difficulty with hydrogen exchange experiments arises during the analysis of the protein and peptide segments. It is critical that once the reaction is quenched or stopped that the protein does not undergo reverse exchange prior to or during its mass spectral analysis. In practice, the samples are subjected to moisture during the ionisation process that can result in deuterium loss. Therefore some care in performing the mass spectral analysis is required to ensure that the level of deuterium in

Biological Mass Spectrometry

127

D,O buffer

HPLCIMS

Deuterium Content Figure 7.9 Typical procedure employed in hydrogen r.x-chunge mass spectrometry experiments to study protein structure (Source: Z. Zhang and D.L. Smith, Protein Science, 1993,2, 522-531)

regions of the protein are not underestimated. One way to overcome this is to employ a reverse-exchange strategy. Here all the amide hydrogens within a protein are first completely exchanged with deuterium. This is usually achieved by denaturing the protein. The level of deuterium incorporation is then verified by mass spectrometry. The protein is then returned to its native state (usually by pH or temperature adjustment) and the reverse-exchange of deuterium with hydrogen monitored in the same manner as that described above. Hydrogen/deuterium exchange has also been used to investigate the structure of proteins and their transient intermediate states in vacuo following the trapping of their ions for extended periods. These studies allow the conformational characteristics and dynamics of gaseous ions to be explored and compared with solution state observations. At least six different intermediate states for the protein, cytochrome c, have been characterised in such gas-phase experiments. 7.2.4.2 Ion Mobility Mass Spectrometry Mass spectrometry can also be employed to study the conformational characteristics of gas phase ions by measuring their mobility as they pass through an inert buffer gas. In these ion mobility measurements, protein ions have drift times that depend upon their average cross-section and hence conformation.

128

Chapter 7

Different drift times have been recorded for the ions of proteins in their native (oxidised or disulphide-bridged) and denatured (reduced) forms.

7.2.4.3 Radical-Based Studies of Protein Structure A recent development that has been employed to study the structure and dynamics of proteins and their interactions by mass spectrometry involves their reaction with radicals. Although the reaction of radicals with proteins in vivo leads to their structural degradation and aggregation through cross-linking, it has been found that proteins can undergo limited oxidation without structural change when the reaction times are kept very short (several milliseconds). Furthermore, the degree to which oxidation occurs is highly dependent on the accessibility of amino acid side chains to the bulk solvent. By measuring the site and degree of oxidation at amino acid markers throughout the protein by mass spectrometry, a protein’s structure can be verified or even predicted. The approach has a number of advantages over hydrogen exchange experiments in that the experiments can be performed extremely rapidly, the radical-induced oxidation reactions are irreversible, and the reaction timescale is sufficiently short to allow some protein conformational changes to be followed at both a global and local level (Figure 7.10). The approach has also been applied to study the dynamics of protein folding and the interactions of proteins with other molecules.

7.2.5 Protein Complexes and Assemblies Beyond the indirect approaches described above, the development of ESI-MS for the study of proteins gave rise to early observations in which ions corresponding to intact protein complexes were sometimes detected within the mass spectrometer. Depending on the solution conditions under which the sample is introduced, and those of the mass spectrometer itself, these complexes can reflect either solution-state or non-specific associations. The pH of the solution, temperatures within the ion source, and the degree to which ions are accelerated within and as they leave the ion source, are all of importance for the detection of gas phase protein complexes. The nature of these gas phase protein complexes is a matter of immediate question, since their molecular weights reflect that they typically have no molecules of solvent attached. Electrostatic interactions between charged groups rather than hydrophobic interactions between the component ions and the solvent are believed to play a more

Biological Muss Spectrometry

129

12-

A Helix: 1-16 7Trp, "Trp

100 8-

06-

0.10 2-

1

10

0 0-0.2

0

' *-

1

2

3

4

B / C Helix: 17-42 24His, 33Phe,36Phe, 37Pto

5

m m

6

* m

--15

1.00.8 -

z

0 6-

.C(

0 4-

,\" 0 2-

__

v

m m m

rn -0.2.

1

0

.

1

1

'

I

2

'

I

I

3

4

G Helix: 103-118

Io3Tyr, Ia6Phe, '13His, I16His

'

"

5

'

6

I

4-

Figure 7.10 Unfolding projiles within segments ojupomyoglobin obtained by limited oxidution qf' the proteiii with hydroxyl mdiculs on millisecond timescri1c.snt increasing concentrations oj'urea (Source: S.D. Maleknia and K.M. Downard, Euv. J. Biochem., 2001,268, 5578-5588, Figure 6)

130

Chapter 7

important role in their stability. It cannot therefore be assumed, even when a known solution-based protein complex is observed in the gas phase, that the complex maintains the same structural features. Nonetheless, a growing body of data now suggests that relationships between solution state and gas phase protein complexes do exist in some systems, and that their direct detection by mass spectrometry may provide a useful means by which to identify such associations in the first instance. This has important ramifications for studies in proteomics, an ultimate description of which requires all protein associations within a cell to be identified. Although the majority of proteins, and other macromolecular complexes, have been detected using ESI mass spectrometry (Figure 7.1 l), some complexes and aggregates have also been detected using the MALDI approach. This is even more surprising since samples are introduced into the mass spectrometer from a solid surface onto which the analyte is added to a high concentration of an organic matrix. Such conditions are expected to dissociate any protein or macromolecular complex prior to the ionisation event. A number of protein and other macromolecular complexes, however, have been both preserved and detected by MALDI mass spectrometry. Among them are immune complexes between protein antigens and monoclonal antibodies (Figure 7.12). 3480

Monomer

Tetrarner

15t pH 6.9

7 6+ 3260

mass

mass 3731

a 74+

b * 0

,

2000

'

3000 m/z

.

.

'

I

4000

5000

Figure 7.1 1 ESI mass spectrum ofstreptavidin in 10 m M ammonium acetate ( p H 6 . 9 ) ut a concentration of 5 pm. Inserts show deconvoluted (by molecular weight) muss spectra at p H 2.5 and 6.9. (Source: J.A. Loo and K.A. Sannes-Lowery, in Mass Spectrometry of Biological Materials, 2nd edn, B.S. Larsen and C.N. McEwen (ed), Marcel Dekker, New York 1998, p. 358, Figure 5 )

Biological Mass Spectrometry

I ~~~~

I

l

2500

15w

24670.0

131

ji

2+ 73993.9

9+

100

I

Figure 7.12 MALDI mass spectra of the tryptic digest of the viralproteins from a type A influenza strain ( a ) before, and ( b ) after interaction with a monoclonal antibody raised to one of the proteins (Source: J.G. Kiselar and K.M. Downard, . I Am. Soc. Mass Spectrom., 2000,11, 746-750, Figure 1)

An antigenic peptide that represents a surface domain of the hemagglutinin antigen of a type A influenza strain is found to preferentially bind to the antibody. Figure 7.12 shows the MALDI mass spectrum for the intact antibody before and after treatment with a mixture of peptides containing the hemagglutinin epitope. The molecular weight change that is evident in the additional ion signals associated with the antibody-peptide complex corresponds to the peptide antigen (MW 2210 Da).

132

Chapter 7

A computer algorithm, known as COMPLX, has been recently developed which enables bimolecular protein and other macromolecular complexes to be identified in an automated manner from both ESI and MALDI mass spectra of the type shown in Figure 7.12. The program is of particular value for spectra that exhibit many hundreds of ion signals such that a manual interpretation of the data would be difficult or impossible. The ability to preserve protein complexes within a mass spectrometer in some circumstances is of potential value for identifying such associations in biological extracts. Beyond such direct observations, a number of indirect approaches have made use of mass spectrometry to study protein complexes. Among these are limited proteolysis of protein complexes to release non-binding domains, hydrogen exchange studies, and the use of microfilters, chromatographic and native electrophoretic methods to separate or isolate large macromolecular complexes for their further characterisation by mass spectrometry.

7.2.6 Proteomics The development of mass spectrometry for the study of proteins has led to its central role in proteomics. Proteomics, or proteome analysis, involves the identification and characterisation of the entire complement of proteins expressed in a single cell or tissue at any given point in time. Functional proteomics seeks to identify the protein components that are unique to diseased cells or tissue, those produced only in response to a genetic abnormality, or those of importance to a particular biological event or process. Two primary methods are used to partially resolve the protein complement of a cell or tissue prior to mass spectrometric detection. The first of these is multi-dimensional chromatography in which proteins are partitioned according to their ionic character, molecular identity, size, and hydrophilicity using ion exchange, affinity, molecular exclusion and reverse-phase chromatography. These partitioned proteins, or their proteolytic products, are then introduced directly into an EST-based mass spectrometer where they are characterised by molecular weight and sequence. A second alternative strategy utilises two-dimensional gel electrophoresis to separate and display the proteins. Proteins isolated from the cell or tissue are separated in 2D gel electrophoresis in the first dimension according to their charge and in the second dimension by size. The proteins are typically digested in-gel to release their proteolytic peptides. These components are analysed by mass spectrometry in both MS and

Biological Muss Spectrometry

133

Figure 7.13 A typical mass spectrometry ( M S ) -based proteomics experiment. Proteins recoveredjiorn cells are partially sepuruted by SDS-PA GE, excised or digested in-gel, and the proteolytic peptides anulvsed on a tandem muss spectronzetev. A Q-TOF hybrid tandem muss spectrometer is shown here (see Chapter 4, Section 4 . 7 ) (Source: adapted from R. Aebersold and M. Mann, Mass spectrometrybased proteomics, Nature, 2003,422, 198-207)

MS/MS experiments (Figure 7.13). Either an ESI or MALDI-based mass spectrometer can be used in these experiments. Proteins are identified by mass spectrometry in the same manner as for single proteins. However, the use of protein and nucleotide databases are of particular importance to these investigations in order that previously characterised proteins, or their homologues, are rapidly identified. Mass map data as well as tandem MS/MS spectra of proteolytic peptides can be used to search a database for a known protein. Where tandem mass spectra are used, algorithms such as SEQUEST are capable of comparing a fragment ion profile with those for any hypothetical proteolytic peptide obtained from a known protein's sequence.

134

Chapter 7

A major challenge confronted in proteomics is the management of large amounts of diverse data. Proteome analysis requires that data collected for each protein from within a cell or tissue be collated in terms of its recovery, electrophoretic or chromatographic profile, biochemical treatment, mass spectral appearance and bioinformatics discovery. Advanced computer-based bioinformatics software and systems are used for this purpose. The field of proteomics has rapidly advanced due to the high performance and continual development of today’s mass spectrometers. The discovery of new, previously uncharacterised proteins is continuing at a rapid rate. Beyond their identification at the molecular weight and sequence level, the quantitation of protein components in cellular lysates and biological extracts is also a core goal of proteomics. This represents a greater analytical challenge but one that has been addressed in part based on chromatographic detector responses, the image analysis of stained proteins on two-dimensional gels, and by mass spectrometry. The chemical treatment of samples isolated from normal and diseased cellular extracts has been used to quantitate the relative levels of protein in each sample. In one protocol, an isotopically-enriched tag is reacted with the cysteine residue side-chains of one sample, while the same unlabelled reagent is reacted with those of the second. The two samples are then mixed and affinity chromatography is used to recover the tagged proteins. Mass spectra are then recorded for the recovered proteins where two ion signals (for the labelled and unlabelled forms) are mass resolved and detected. The relative area for these ion signals provides a quantititative measure of the levels of protein in each sample (Figure 7.14). Disadvantages of such an approach include the need to couple the reagant to proteins in each sample with equal efficiency and the ability to recover the proteins by affinity chromatography. A particularly powerful application of MALDI-MS in proteomics is that involving a direct analysis of the spatial organisation (to a resolution of some 50 pm) of peptides and proteins in mammalian tissue sections. Caprioli and colleagues have demonstrated that it is possible to laser ablate compounds directly from a tissue section, or a blot of the tissue slice, to produce ions that span the range from mIz 1,000 to 100,000. Two-dimensional maps can then be reconstructed based on the intensity of these ions to provide, to a first approximation, the relative levels of certain molecules within the tissue (Figure 7.15). This mass spectrometric “imaging” of proteins and other components from tissue is still in its infancy but if well-developed would provide a powerful and rapid technique for use in both research and clinical settings.

Biological Mass Spectrometry

135

Figure 7.14 The isotope-coded aflnity tag (ICAT) approuch to quantitate the relative levels ofproteins from cellular extracts by mass spectrometry (Source: adapted from S.P. Gygi, R. Aebersold and M. Mann, Mass spectrometry and proteomics, Cum Opinion in Chem. Biol., 2000, Vol. 4, 489494, Figure 2)

In a similar manner, MALDI mass spectrometry has been used to obtain protein profiles of unfractionated microorganisms including viruses, bacterial and fungal cells, and spores. The positive and negative MALDI mass spectra of proteins desorbed directly from the bacteria Helicobacter pylori, where 26995 intact cells were introduced into the mass spectrometer, are shown in Figure 7.16. Such spectra have also been generated using laser ablation mass spectrometry where airborne microorganisms are introduced into the mass spectrometer as an aerosol. Microorganisms may also be identified, based on an analysis of the proteolytic peptides generated from their proteins in a mass map experiment. A future goal of proteomics is to characterise the association of proteins en rnasse in cells and tissues. This will allow studies of

136 (a)

Chapter 7

loo

7059

7532

6844. (d)

5890 60

1

1

6°62

5632, (c) 14117

,

11 297

2000

6600

11 338,(e) I12122

1 1 200

15 800

20 400

25 000

Mass (m/z)

Figure 7.15 MALDI-MS imaging oj'u trunsverul section of rut bruin. (a)Survey profile taken randomly ucross the section, ( b ) opticul imuge of the section before matrix application, (c)-(gj ion density maps obtuined at diflwent m / z vulues (Source: P. Chaurand, S.A. Schwartz and R.M. Caprioli, Imaging mass spectrometry: a new tool to investigate the spatial organization of peptides and proteins in mammalian tissue sections, Current Opinion in Chetnicul Biology, 2002, 6(5), 676-68 1 , Figure 3)

Biological Mass Spectrometry

137 xi0

7682

! 80 -

40

-

89721

7680 5527

9113

8322

1

10393

9620

1 6000

25000

Masslcharge

9110

20 4000

20000

8000

I

10000

Masslcharge

Figure 7.1 6 Positive and negative M A L D I Mzuss spectra of proteins desorbed directlyfjom the bacteria Helicobacter pylori (Source: PA. Demirev, J.S. Lin, F.J. Pineda and C. Fenselau, Bioinformatics and mass spectrometry for microorganism identification: Proteome-wide post-translational modifications and database search algorithms for characterization of intact H. pylori, Anal. Chern., 2001,73(19), 45664573, Figure 1)

protein function to be followed on a global scale. Current genetic and biochemical methods are sure to be supplemented by the use of mass spectrometry for these endeavours, particularly since mass spectrometry is already in widespread use for the study ofprotein structure and interactions. The use of MALDI mass spectrometry in this regard has been shown to be able to survey the structure and antigenic identity of the influenza virus. This has been achieved without the need to immobilise either the viral antigens or their cross-interacting antibodies. It has been found that the interaction between an antibody and a specific region (epitope) of a protein antigen can be preserved on a MALDI surface from which all non- binding peptides, generated after proteolysis, can be preferentially ionised. A careful comparison of the MALDI mass spectrum of the proteolysis products of an unreacted antigen mixture (the control) versus that of the antibody-containing mixture enables binding domains to be identified and characterised. Such spectra for the tryptic digest of all four viral antigens from a type A influenza strain are shown in Figure 7.17. A measure of the ion abundances in both spectra enable an

138

Chapter 7

i

lob0

500

lD00

i

ibw

zoo0

2500

M8M (mJ4

Figure 7.17 MALDI mass spectra of the tryptic digest ofthe viralproteins jrorn a type A injluenza strain ( a ) before and ( b ) after reaction with monoclonal untibody. The peak labelled with an asterisk is attributed to u matrix cluster ion (Source: J.G. Kiselar and K.M. Downard, Biochemistry, 1999,38, 14185-14191, Figure 2)

antibody-binding domain (mlz 2,210.1) to be identified within a single (hemagglutinin) antigen based upon its reduced intensity. 7.3 OLIGONUCLEOTIDES AND NUCLEIC ACIDS Like peptides, oligonucleotides are linear polymers comprised of nucleoside monomers linked by a phosphodiester group. Conveniently, these polymers are composed of just four different natural monomer units: adenosine (A), guanosine (G), cytidine (C) and uridine (U) in the case of ribonucleic acids (RNA), and the deoxy forms of A,G,C and thymidine

Biological Muss Spectrometry

139

(T) in the case of deoxyribonucleic acids (DNA) (see Appendix 9). This would appear to make the identification and sequencing of nucleic acids by mass spectrometry more straightforward than comparably sized proteins. In reality, however, oligonucleotides and nucleic acids are more fragile within a mass spectrometer than proteins and prone to degradation at the phosphodiester linking group. The larger mass of each of the mononucleotides also impacts the ionisation efficiency of oligonucleotides and nucleic acids and the ability to induce their dissociation in tandem mass spectrometry experiments. Nonetheless, mass spectrometry has been applied to the molecular weight and sequence analysis of oligonucleotides and nucleic acids. 7.3.1 Identification of Modified Nucleosides Post-transcriptional and other cellular processes result in a wide range of structural modifications occurring at the purine and pyrimidine bases of RNA. Mass spectrometry has played a significant role in the identification of these modified bases, originally through the enzymatic hydrolysis of RNA. Such hydrolysates are analysed by either LC-ESI-MS, or GCMS in the form of their volatile trimethylsilylated derivatives. The mass difference between the nucleoside and natural forms provides information concerning the type of modification(s). Tandem (MS/MS) mass spectrometry of the modified nucleosides can also be employed to enable the site of the modifications to be determined. Although both RNAand DNA segments in excess of 100bases have been successfully detected within mass spectrometers, they are typically ionised with less efficiency than proteins of a comparable size. This is associated with the fact that the negatively charged phosphodiester group shows both a propensity to dissociate and to adduct alkali metal cations such as Na' and K'. These traits adversely impact the study of the low amounts of DNA and RNA that is usually available. As a consequence, mass spectrometry is yet to compete with some other analytical approaches for the detection and characterisation of nucleic acids. It does, however, offer a highly complementary approach and has proved useful for certain applications.

7.3.2 Sequencing of Oligonucleotides by Tandem Mass Spectrometry High-resolution mass spectrometry can be used to verify the sequence of a synthetic or isolated RNA and DNA segment. To avoid contributions to their mass from metal ions, samples are desalted prior to analysis.

140

Chupter 7

Like proteins, RNA and DNA can be digested into smaller oligonucleotides that are more amenable to analysis by the use of restriction endonucleases. These segments and synthetic oligonucleotides can be sequenced in the same manner as described for peptides (:see Section 7.2.3.2). Tandem mass spectrometry has been applied to sequence simple short-chain oligonucleotides and a nomenclature has been proposed to describe the observed fragments (Figure 7.18).

\OH Figure 7.18 Nomencluture to designate fragment ions detected upon the dissociution of oligonucleotides in a muss spectrometer

The ESI tandem CID spectrum of oligonucleotide d(GCTGGCATCCGT) recorded in the negative ion mode is shown in Figure 7.19. Although MS/MS experiments are useful for particular applications, they fail to compete with automated high-throughput sequencing by means of the polymerase chain reaction (PCR).

7.4 OLIGOSACCHARIDES AND GLYCOCONJUGATES Oligosaccharides and glycoconjugates are composed of monosaccharides such as glucose, hexose and fructose linked through glycosidic bonds. Determining the structure of these compounds is far more difficult than is the case for peptides and oligonucleotides. This is due to the isomeric nature of some monosaccharides and the fact that these monomers can couple in a multitude of ways to produce highly branched structures. Like oligonucleotides, oligosaccharides and glyconjugates also ionise less efficiently than their equivalently sized peptide and protein counterparts. It is often necessary to analyse oligosaccharides in the negative ion mode due to the propensity of free hydroxyl groups to support a negative charge. Alkali metal ions often present in a sample may coordinate to these groups resulting in an unpredictable increase in the mass of the compound. To prevent this, glyconjugates can be methylated, acetylated or otherwise derivatised to facilitate their ionisation as positive ions

Biological Mass Spectrometry

141 (4

44 GCT GGC ATC CGT

12

M,=3814

[M-3H-GI3W,p 'c1

12.)

1

1187.1

11

125.3

'I-.-

[M-3HI31295.1

1270.0

I

I

Figure 7.19 ESI tandem muss spectrum of the [M - 3H]'- ion ojoligonucleotidt. d( GCTGGCA TCCGT) from which its sequence can be derived (Source: R.H. Griffey, M.J. Greig, H.J. Gaus, K. Liu, D. Monteith, M. Winniman, L.L. Cummins, Characterisation of oligonucleotide metabolism in vivo via liquid chromatography/electrospray mass spectrometry with a quadrupole ion trap mass spectrometer, J. Mass Spectrom., 1997,32, 305-3 13.)

''

[M+2 N a] -

f

1116.674

1 1 30,691

2+ 1119.692

I

1120

2+

1j23.676

1130

1140

m/r

Figure 7.20 Moleculur ion region of the ESI FT-ICR muss spectrum of permethyluted gang1ioside G, ,, (Source: C.E. Costello, Bioanalytic applications of mass spectrometry, Current Opinion in Biotechnology, 1999, 10(1), 22-28, Figure 2a)

,

Chapter 7

142

(Figure 7.20). This method of analysis proved mandatory prior to the development of the newer desorption ionisation methods (such as FAB and MALDI) in order to volatilise oligosaccharides. Heterogeneity is often encountered in glyconjugate samples where individual components differ in molecular weight by that of a monosaccharide (Figure 7.21). A molecular weight profile of the sample provides an immediate indication as to its complexity and may identify potential heterogeneity among the components. Hydrolysis of the sample prior to analysis can allow the component monosaccharides to be identified though care must be taken to avoid modifying groups being removed in this process. Glycosidases are also used to evaluate such samples as described later in section 7.4.2.

ml. '94

[(A-G4S)3]3ndz 391

\

[(A-G4S),(A-G)13m l z 493 [(A-G4S),(A-G)12ndz 547

ndz 789

9 400

800

1200

1600

2000

2400

mlz Figure 7.21 ESI mass spectrum of the enzymic Jgydrolysate oj'kappa-carrageenan containing ions associated with the tetrasaccharide ( A - G4S)2and hexasaccharide (A-G4S), (Source: D. Ekeberg, S.H. Knutsen and M. Sletmoen, Negative-ion electrospray ionisation-mass spectrometry (ESI-MS) as a tool for analysing structural heterogeneity in kappa-carrageenan oligosaccharides, Carboyhydrate Research, 2001,334(1), 49-59, Figure 4)

7.4.1 Sequencing of Oligosaccharides by Tandem Mass Spectrometry

The dissociation of oligosaccharides within a tandem mass spectrometer results primarily in the cleavage of the glycosidic bonds. A nomenclature

Biological Muss Spectrometry

143

has been proposed to designate the possible fragment ions formed by the dissociation about glycosidic bonds in which charge is retained at either the reducing or non-reducing terminus resulting in the production of B, C, Y and Z ions (Figure 7.22). A numbered subscript identifies the number of monomeric units toward the termini where the B and C ions contain the non-reducing terminus and Y and Z ions contain the reducing terminus. The Greek letters (a,p, etc.) are used to denote the branch position by a subscript. Two additional fragments A and X are a result of ring cleavage. These ions are designated with numerical superscripts to denote the bonds broken within the ring. For instance, the ion nomenclature ‘q3A2,denotes a fragment formed from the cleavage of the first and third bonds of the monosaccharide unit (in a clockwise direction from the 0 - C bond, denoted bond zero) at the first two sugars from the non-reducing terminus along the first (a)branch.

I

OH

25A3

Figure 7.22 Nomenclature to designate the fragments ionsformed by the dissociation of oligosaccharides in a mass spectrometer (Source: J. Vdth and C.E. Costello, in Methods in Enzymology, McCloskey (ed), Academic Press, New York, 1990 Vol. 193, Ch. 40, p. 743, Figure 2)

As for peptides and oligonucleotides, the sequence and structure of oligosaccharides can be assembled from the mlz values of their fragment ions. Appendix 10 shows the mass increments for common monosaccharide units, representing the molecular weight of a monosaccharide less 18 u for a molecule of water. To illustrate this process, the tandem mass spectrum of a pentasaccharide recorded in the negative ion mode is shown in Figure 7.23. Fragment ions that result from glycosidic bond cleavages are evident in this spectrum in addition to those formed by ring cleavage (A and X ions). The appearance of abundant 2,4A3and Y,, ions at mlz 545 and 586 enables the branched nature of the structure to be deduced.

144

Chupter 7

2.4A

545

I Y3a 3P

1

'O;

c,

74x2

179

262

Mass

Figure 7.23 Negative ion CID tandem muss spectrum of. heptasacchari~ltrfrom N-linked glycum (Source: D.L. Gillece-Castro and A.L. Burlingame, in Methods in Enzymology, McCloskey (ed), Academic Press, New York, Vol. 193, Ch. 37, p. 689, Figure 3)

7.4.2 Exoglycosidase Digestion An alternate strategy to tandem mass spectrometry is to digest the oligosaccharide with a series of exoglycosidases and record the molecular weights of the resulting products. A number of commercial exoglycosidases are available that show a range of specificities for oligosaccharide cleavage. Sialidases, galactosidases and mannosidases can be employed to cleave glyosidic bonds at specific monosaccharides thus releasing these components. Figure 7.24 represents in cartoon form how glycosidases can be used to distinguish between alternate biantennary and triantennary structures by MS analysis after stepwise digestion. 7.4.3 Derivatisation Approaches: Oxidative and Reductive Cleavage to Identify Branching

There are many occasions where it is desirable, even necessary, to derivatise an oligosaccharide prior to its analysis. This is conducted

Biological Mass Spectrometry

145 Triantennarv

Biantennarv

Remo

** I

Remove PI-2 GlcNAc

Remove al-3 Man

Remove P I 4 GlcNAc Key: pl4Gal @

4a16Man

Remove 1x1-3Man

p 1-3 Gal

Ip 1-2 GlcNAc

p 1-4 GlcNAc

O‘

a1-3Man

0-

p 1-4 Man

Figure 7.24 Exoglycosiduse digestion used to determine the structure of brunched oligosucchur ides (Source: adapted from R. Orlando and Y. Yang, in Muss Spectrometry of Biological Materials, 2nd edn, B.S. Larsen and C.N. McEwen (ed), Marcel Dekker, New York, USA, 1998, Ch.9, p. 236, Figure 11)

in order to improve the volatility of the molecule, increase the yield of parent ions, direct its fragmentation in a more predicatable manner or assist with the investigation of particular structural features. Methylation and acetylation reactions are among the most common derivatisation reactions employed. These reactions, however, show little specificity for functional groups within the sugar and as such are generally used to completely derivatise all such groups (known as permethylation and peracetylation) to improve the molecule’s ionisation efficiency in the positive ion mode. Periodate oxidation provides a more selective strategy for probing structural elements within an oligosaccharide. The reaction involves the cleavage of C-C bonds and allows for the branching pattern within 0linked oligosaccharides to be determined. This is illustrated in Figure 7.25 where a 1,4-linked and 1’6-linked hexose unit undergoes ring opening following oxidation and reduction to yield monosaccharideunits of different mass (208 and 164 u respectively) after methylation. The same chemistries can be used to determine the anomeric configuration of glycosidic bonds by mass spectrometry. It is possible to

146

Chapter 7 OH

I

HC-CH

/

0

OH

\

I

II

104-

0

HC-CH

I1

OH

NaBD4 ---+

.......................

.......................

r

I ,HC(D)

OMe

OH OMe methylation (D)HC, ,HC(D)

-

1

............................

I

\

............................

1 OMe

-0-CHz

6 R + 208u

I

(D)HC

1

' R + 164u

Figure 7.25 IdentlJication oJ'1,4-linked and 1,6-linked hexose after ring opening by per iodate oxidation, reduct ion and me thylat ion

selectively oxidise the j3-anomer of hexose, over its a counterpart, leading to a mass shift being detected in this region of the oligosaccharide (Figure 7.26).

-04--+p CH20Me

OMe

OMe

CH20Me Cr03

____)

c +

-o$o

OMe

OMe

Figure 7.26 Oxidisation of the p-anomer of hexose, that occurs selectively over its a counterpart, can be identified by a mass sh$t

FURTHER READING G. Siuzdak, Muss Spectrometry for Biotechnology, Academic Press, 1996. K.M. Downard Advances in protein analysis and sequencing by mass spectrometry, New Adv. Anal. Chem., 2000, P2, 1-30. A.G. Marshall, C.L. Hendrickson and S.D.H. Shi, Scaling MS Plateaus with FTICR MS, Anal. Chem., 2002,74,252A-259A. W.J. Henzel, T.M. Billeci, J.T. Stults, S.C. Wong, C. Grimley and C. Watanabe, Identifying proteins from two-dimensional gels by molecular mass searching of peptide fragments in protein sequence databases, Proc. Nutl. Acad. Sci. USA, 1993,90, 501 1-5015.

Biological Mass Spectrometry

147

J.R. Yates 111, S. Speicher, P.R. Griffin and T. Hunkapiller, Peptide mass maps: A highly informative approach to protein identification, A n d . Biochem., 1993,214, 397. M. Mann, P. Hojrup and P. Roepstorff, Use of mass spectrometric molecular weight information to identify proteins in sequence databases, Biol. Mass Spectrom., 1993,22, 338-345. K. Biemann and I.A. Papayannopoulos, Amino acid sequencing of proteins, Acc. Chem. Res., 1994,27(11), 370-378. I.A. Papayannopoulos, The interpretation of collision-induced dissociation tandem mass spectra of peptides, Mass Spectrom. Rev, 1995,14(1), 49-73. J.R. Engen, D.L. Smith, Analysis of proteins with hydrogen exchange and mass spectrometry, Anal. Chem., 2001, 73, 256A-265A. S.D. Maleknia, K.M. Downard, Radical approaches to probe protein structure, folding, and interactions by mass spectrometry, Mass Spectrom. Rev., 200 1, 20(6), 388-40 1. J.A. Loo, Studying noncovalent protein complexes by electrospray ionization mass spectrometry, Mass Spectrom. Rev., 1997, 16(1), 1-23. J.G. Kiselar and K.M. Downard, Antigenic Surveillance of the Influenza Virus by Mass Spectrometry, Biochemistry, 1999, 38(43), 14185-1 4 191. M.R. Wilkins, K.L. Williams, R.D. Appel and D.F. Hochstrasser, Proteome Research: New Frontiers in Functional Genomics, Springer Verlag, Berlin, Germany, 1997. J.A. McCloskey, A.B. Whitehill, J. Rozenski, F. Qiu and P.F. Crain, New techniques for the rapid characterization of oligonucleotides by mass spectrometry, Nucleosides & Nucleotides, 1999, 18, 1549-1553. B. Reinhold, V. Reinhold and C.E. Costello, Carbohydrate molecular weight profiling, sequence, linkage and branching data: ESI-MS and CID, Anal. Chem., 1995,67,1772-1784. K.H. Khoo and A. Dell, Assignment of anomeric configurations of pyranose sugars in oligosaccharides using a sensitive FAB-MS strategy, Glycobiology, 1990, 1 83-9 11.

CHAPTER 8

Mass Spectrometry in Medical Research 8.1 CHARACTERISATION AND QUANTITATION OF DRUGS AND METABOLITES 8.1.1 Introduction In recent years, there has been an appreciable shift in focus in drug discovery away from the design, synthesis, characterisation and evaluation of organic drug molecules. Today’s pharmaceutical and academic research laboratories seek to gain a global understanding of the genetic basis of disease states (through functional genomics), potential causative proteins or markers (functional proteomics), and the evaluation of drug candidates and therapies. Pharmacological studies of drug absorption, excretion and metabolism are also performed in the context of a complete description of human biology (metabolomics). The identification and characterisation of these biomarkers or targets can be performed by mass spectrometry using approaches described in Chapter 7. The traditional characterisation and evaluation of organic drug targets, nonetheless, continues to be an important area of medical research. In these studies, molecules are designed to “target” a biological molecule, tissue or system to stimulate, confer or prevent some function or activity. Such molecules are usually constructed by synthetic routes either as a single lead compound or as part of a chemical library of related compounds. Potential drug targets are assessed based upon their bioavailability (or the extent to which the administered dose reaches its target), half-life, and therapeutic index. The half-life of the compound represents the time in which it takes for 50% of the concentration of the compound to be excreted or metabolised in vivo. The therapeutic index represents a measure of the desired function or activity versus any undesired side effects. The most effective drugs are those with a high therapeutic index, a high bioavailability, and often a low half-life.

148

Muss Spectrometry in Medical Research

149

8.1.2 Sample Preparation Techniques in Drug Discovery When dealing with biological and in vitvo samples, the preparation of the sample for mass spectrometric analysis is of paramount importance in order to achieve analytical success. A wide variety of approaches including precipitation and centrifugation methods, ultrafiltration, solid-phase extraction, blotting and immo bilisation are all employed subject to the sample at hand and its biological source. The precipitation of biological components from an extract can be achieved by adding denaturing solvents such as methanol, acetonitrile or acetone. The recovery of these components follows the rapid mixing of the solution and their density centrifugation to the base of the sample vial. Ultrafiltration provides another means with which to separate the components of biological mixtures by passing the solution through molecular weight cut-off filters. Under centrifugation the larger molecular weight species remain trapped on the top of the filter while smaller compounds pass through with the solvent. This process can be performed in a series of stages so that a sample mixture is effectively partitioned into sets of compounds spanning several molecular weight ranges. Solid-phase extraction (SPE) takes advantage of the separation characteristics exploited in liquid chromatography. Samples are loaded onto small cartridges, pre-packed with chromatographic supports suitable for reverse-phase, ion exchange or ion exclusion chromatography. The sample solution is passed through the cartridge under gravity, by vacuum or using centrifugal force. Solvents of differing compositions are then added in succession to effect the partitioning of solutes between the solid and solution phases. Blotting and immobilisation procedures are also used widely in drug discovery applications to isolate particular components from biological sources. Drug targets can also be absorbed or immobilised on films and screened against an array of drug compounds ahead of their analysis. The screening of drugs using mass spectrometry is the subject of Section 8.4. 8.1.3 Qualitative Analysis of Organic Drugs and their Metabolites The identification or characterisation of a drug compound by mass spectrometry is dependent upon the structural features of that compound. For most small to moderately sized (-1000 Da.) compounds this is achieved by either GC-MS or LC-ESI-MS. The former approach predated the development of the ESI technique and has to some extent been superseded by it. However, certain volatile or derivatised molecules are particularly suited to GC-MS in which the resulting EI (or CI) mass spectrum provides both molecular weight and structural information.

Chapter 8

150

The latter is usually achieved through the use of tandem mass spectrometry in LC-ESI-MS experiments since the “soft” nature of the ESI technique results in most compounds resisting fragmentation. Such approaches can be used for the characterisation of trial drugs, those approved for pharmaceutical use, or illegal drugs or narcotics. As an illustration, cocaine and 6-acetylmorphine can be detected in a single human hair by GC-MS. A cutting from a single hair was washed and heated in methanol. The volatile components were passed into the ion source of a GC-MS and both cocaine and 6-acetylmorphine were detected by means of selected ion monitoring (SIM) (Chapter 5, Section 5.4.2) of the fragment ions at mlz 182 (Figure 8.1) and 268 respectively. S I M (mlz182)

Cocaine

A) Cocaine User

0

1

2

J

3

4

5

6

Time (mint

Figure 8.1 SIM of the mlz 182 fragment ofcocaine from: ( A ) a I-cmpiece of a cocaine user’s hair (subject 2 ) , ( B ) The rerun of the same hair sample used in ( A ) demonstrating the near 100% recovery, and ( C ) a I -cm piece oj’hair obtained from a drug free individual (Source: S.B. Wainhaus, N. Tzanani, S. Dagan, M.L. Miller and A. Amirav, Mass Spectvom., 1998, Fast Analysis of Drugs in a Single Hair, .J Am. SOC. 9, 131 1, Figure 4)

Mass Spectrometry in Medical Research

151

OH

t

7’

“3 404

e67

i

Figure 8.2 Particle beamlchemical ionisation datafor. ( a ) total ion current (mlz 700-1300) chromatogram of rut urine, ( b ) mass spectrum of component “A” the metabolite eluting at 12.6 min., and ( c ) mass spectrum of metabolite rejerence standard (Source: adapted from L. Iavarone, M. Scandola, F. Pugnaghi and P. Grossi, Qualitative Analysis of potential metabolites and degradation products of a new antiinfective drug in rat urine, 1 Pharma. Biumed. Anal., 1995, 13, 607-614, Figure 8)

These constituents were detected at concentrations as low as 10 partsper-billion (ppb) with analyses conducted within 10 minutes. The qualitative analysis of metabolites is also an important requirement of drug testing to evaluate both the half-life of a drug and the nature and potential toxicity of its metabolites. LC-MS has been employed to follow the metabolic fate of a p-lactam antibiotic. The metabolic profiles of the antibiotic in rat urine were monitored by mass spectrometry in conjunction with ion exchange HPLC after a single intravenous administration of the drug. Two

Chapter 8

152 192s

Figure 8.3 Deconvoluted (molecular weight adjusted) ESI mass .rpectrum oj'a 1: 1 ras GDP oncoprotein inhibitor ( S C H 54292) complex with an average mass of1 9,816 Da (Source: A.K. Ganguly, B.N. Pramanik, E.C. Huang, S. Liberles, L. Heimark, et al., Detection and structural characterisation of Ras oncoprotein-inhibitors complexes by electrospray mass spectrometry, Bioorganic and Medicinal Chemistry, 1997, 5(5), 817-820, Figure 1)

metabolites were detected, one of which corresponds to a ring-opened degradation product based upon its mass spectrum (Figure 8.2). The association of drugs with protein targets can also be evaluated by mass spectrometry. As described in Chapter 7 (Section 7.5.2), ESI and to a lesser extent MALDT mass spectrometry are able to both preserve and detect specific solution state associations within a mass spectrometer. A 1:l non-covalent complex between a ras GDP oncoprotein and a potential inhibitor is shown in Figure 8.3. Mutant ras proteins have been implicated in the growth of a wide range of human tumours and thus the inhibition of GDP could prevent continued tumour growth. The approach offers the opportunity to study protein complexes that cannot be investigated by other methods. In this case, crystals for the ras G D P protein could not be successfully prepared for X-ray crystallography. The purity of a drug compound must also be thoroughly evaluated before it can be used in clinical trials. In some cases it is necessary that the drug be enantiomerically pure. Where a chiral drug has a particular potency and the drug is to be administered as a racemic mixture, it is necessary to establish that the inactive enantiomer affords no potential side effects.

Muss Spectrometry in Medical Research

153

The kinetic method developed by Cooks (Section 6.1.4) has been used to determine the enantiomeric composition of a drug mixture through the competitive dissociation of their copper-bound ion complexes. Copper(I1)-bound ion complexes formed from seven model drugs together with a series of chiral reference compounds (L-amino acids only) were analysed by electrospray ionisation mass spectrometry. The complexes were found to undergo collisionally activated dissociation (CAD) by competitive loss of either the neutral drug molecule or the reference. The ratio of the two competitive dissociation rates allowed the composition of enantiomeric drugs in the mixture to be determined using a two-point calibration curve. 8.1.4 Quantitative Analysis of Drug Compounds and their Metabolites Mass spectrometry plays a central role not just in confirming the presence or structure of a drug molecule, but also in measuring the absolute and relative levels of the compound or its metabolites in serum or plasma. A measurement of the concentration of the compound or its metabolites as a function of time, after the dose has been administered, provides a pharmacokinetic profile that is useful in establishing a drugs’ bioavailability or its rate of metabolism throughout the body. Quantitation measurements are also necessary where the metabolite is toxic or pharmacologically active when it reaches a particular concentration. In addition to selected ion monitoring (SIM) described above, selected or multiple reaction monitoring (SRM or MRM) is one of the most common approaches used for this purpose and is achieved within a tandem mass spectrometer. Here the metastable transition or conversion of a metabolic precursor to a product is monitored. Only selected precursor ions that decompose to product ions of a particular m/z ratio will be detected. This affords optimal sensitivities where compounds can be quantitated to the sub-ppb (part-per-billion) level. In a MRM experiment, a series of such reactions is monitored by rapid switching of the electric fields applied to the mass analyser to study each reaction in turn. Figure 8.4 shows the MRM ion chromatograms for seven dosed compounds extracted from brain tissue plus an internal standard (top chromatogram).

154

Chapter 8

Figure 8.4 ESIpositive ion M R M chromatograms jor seven compounds extmctedjrom brain tissue plus an internal standard (top chromatogram) (Source: D.T.Rossi, Sample preparation and handling for LC/MS in drug discovery, in Mass Spectrometry in Drug Discovery, D.T.Rossi and M.W.Sinz (ed), Marcel Dekker, New York, 2002, Ch. 6, p. 206, Fig. 15)

8.2 DEFINING METABOLIC PATHWAYS WITH MASS SPECTROMETRY Where the pharmacokinetic profile of a metabolite exists. it is often desirable to establish the reaction pathway through which the meta-

Mass Spectrometry in Medical Research

155

bolite is produced. Common metabolic processes include oxidation, methylation, acetylation and epoxidation as well as degradation reactions. These consequently cause changes in the molecular weight of the product in each step of a metabolic pathway. Heavy isotopes (e.g. I3C, ”N) are of use for tracking the decomposition or reaction of a drug as it is metabolised. The isotopic enrichment of a drug changes its molecular weight and isotopic profile (Chapter 1, Section 1.2). If the heavy isotope is retained by the metabolite its ion will have a mlz ratio and isotopic distribution that will differ in appearance from that of its unlabelled form. Thus the drug and metabolite can be associated. The use of a series of labels at different positions throughout the drug enables a range of metabolites to be identified from which their dissociation pathways can be deduced. The placement of the heavy isotope should be decided upon with some care and it must be placed in a metabolically inert position since if the isotope “lost” early during its metabolism, the reaction pathway can no longer be followed. 8.3 CHARACTERISATION OF DRUG LIBRARIES BY MASS SPECTROMETRY As mentioned at the introduction to this chapter, drugs and drug targets are usually constructed as part of a chemical library. These libraries are pooled and split and subsequent sets of compounds screened for activity in an assay. Library components can be characterised in these sets by reverse phase HPLC or capillary electrophoresis coupled to an ESI or APCI mass spectrometer (Chapter 3, Section 3.2.10). It is common to pass a portion of the eluant from the column to a secondary detector, such as an ultraviolet (UV) absorbance detector preferably operating over a range of wavelengths. The purpose of the secondary detector is to assist with the detection and quantitation of the components. Stable isotopes can also be used to encode particular components of drug libraries to assist with their identification in what is termed stable isotope encoding. In many instances, libraries are constructed of structurally similar compounds whose molecular weights may coincidentally be identical or indistinguishable by mass spectrometry. If stable heavy isotopes are incorporated into some compounds during their synthesis, these can be mixed in series at varying molar ratios with their non-labelled counterparts, such that each component can be distinguished based upon the ratio of the ion signals for the labelled and unlabelled forms. Establishing the chemical components in each library enables screening of the activity of these components to begin. For the most part, this

156

Chapter 8

is performed without the aid of a mass spectrometer but a number of mass spectrometric-based assays have now been developed that are of use in drug discovery investigations.

8.4 DRUG SCREENING USING MASS SPECTROMETRY The immobilisation of a drug target to a surface or membrane provides a means with which to screen drug libraries in an automated manner. A solution of mixtures of library compounds can be passed across the bound target and the surface washed after an appropriate incubation time. The bound drugs can then be chemically released and detected. Mass spectrometry has been employed in such assays to detect the released drug including the use of MALDI to catalyse that release and subsequently ionise the bound drug. A multi-sample MALDI target or miniaturised chip format enables a series of drug interactions to be studied simultaneously where each position on the target or chip characterises a unique association. Such an approach has been used to screen protein associations in what has been described as biomolecular interaction analysis mass spectrometry (BIA-MS) (Figure 8.5).

Figure 8.5 Schematic representation of the BIA-MS method in which surjhcc-imniobilised ligands with afinities toward a protein of interest are used to retrieve the protein from a complex biological mixture. Surjace plasma resonance is used to monitor the interaction and quantijy the amount of retrievedprorein(s). MALDI-MS releases the bound protein enabling it to be identlfied by molecular weight (Source: D. Nedelkov and R.W. Nelson, Biomolecular interaction analysis mass spectrometry: A comprehensive microscale proteomics approach, American Laboratory, 2001, 22-25, Figure 1)

Mass Spectrometry in Medical Research

157

In an extension of this approach, Williams, Nelson and co-workers have used a 96-well robotic workstation to preferentially isolate proteins in parallel from human blood and plasma using tips filled with immobilised antibody to each protein target. The isolated proteins were subsequently transferred to a MALDI sample target for their detection by mass spectrometry (Figure 8.6). The affinity isolation of proteins from such biological matrices in combination with MALDI-TOF MS holds promise in proteomics for the detection of protein markers or mutants associated with genetic abnormalities and disease. 8.5 TRACE ELEMENT ANALYSIS IN NUTRITION

Just as isotopic labelling techniques are used to follow the absorption and metabolism of a man-made drug administered orally or intravenously, mass spectrometry can also be applied to follow the ingestion and absorption of essential vitamins and minerals. Because not all nutrients and minerals in a diet are retained and available for physiological function, the uptake of dietary elements can be followed using stable heavy isotopes. Dietary iron is vital for correct physiological function and a lack of iron leads to anemia, impaired mental and motor development, and a reduced resistance to infection. The absorption of iron by red blood cells can be followed using test meals isotopically-enriched with 57Feor "Fe. Subjects are fed meals containing either 57Feor "Fe and the amount of isotope retained in the blood is quantified by mass spectrometry. Inductively coupledplusmu muss spectrometry (ICP-MS) is used widely for such an analysis (see Section 9.1.1) and employs a high resolution mass analyser and multiple ion detectors. ICP-MS has also been employed to study calcium uptake in bone. During bone growth, calcium is transferred into bone at rates greater than is lost. The balance alters with age resulting in a reduction in bone mineral density and the onset of osteoperosis. Women are at particular risk of osteoperosis and the resultant bone fractures. Stable isotope studies using diets containing 40Ca and 41Caallow the absorption of calcium to be followed by monitoring calcium levels excreted in fecal matter or urine, or present in blood plasma.

Chapter 8

158

A 1

2

3

4

5

1

2

3

4

5

B 6

7

Figure 8.6

8

9 1 0 1 1 1 2

Mass Spectrometry in Medical Reseurch

159

FURTHER READING D.T. Rossi and M.W. Sinz (ed) Mass Spectrometry in Drug Discovery, Marcel Dekker, New York, 2002. D.I. Papac and Z. Shahrokh, Mass spectrometry innovations in drug discovery and development, Pharmaceutical Research, 200 1, 18(2), 131-145. S.J. Gaskell and D.S. Millington, Selected metastable peak monitoring: A new specific technique in quantitative gas chromatography mass spectrometry, Biomed. Mass Spectrom., 1978,5, 557-558. U.A. Kiernan, K.A. Tubbs, K. Gruber, D. Nedelkov, E.E. Niederkofler, P. Williams and R.W. Nelson, High-throughput protein characterization using a mass spectrometric immunoassay, Anal. Biochem. 2002,301,49-56. F. Mellon, R. Self and J.R. Sartin, Muss Spectrometry of Natural Substances in Food, Royal Society of Chemistry, Cambridge, UK.

Figure 8.6 (opposite) High-throughput Mass Spectrometric Immunoassay ( M S I A ) analysis of transthyretin ( T T R ) and retinol binding protein ( R B P ) using human plasma samples from six individuals randomly arranged in a %-well titer plate. ( A ) Mass spectra result from M S I A analysis utilising anti- T T R derivatisedpipette tips. Shown is the region of the singly charged T T R signals. Highlighted cells show spectra with resolvable parent ion diflerences between each other. ( B ) Mass spectra result from the M S I A analysis ojthe same samples utilising anti-RBP derivatisedpipette tips. Shown is the region of the doubly charged RBP signals (Source: U.A. Kiernan, K.A. Tubbs, K. Gruber, D. Nedelkov, E.E. Niederkofler, P. Williams and R.W. Nelson, High-Throughput Protein Characterization Using Mass Spectrometric Tmmunoassay, Anal. Biochem., 2002,301,49-56, Figure 3)

CHAPTER 9

Mass Spectrometry in the Environmental and Surface Sciences 9.1 ENVIRONMENTAL ANALYSIS The mass spectrometer is the most widely used detector to analyse the effects of natural and man-made substances on the environment. Social pressures as well as the need for ecological sustainability have necessitated that changes to our environment are monitored with some precision. The Earth’s growing population has placed increased demands on the planet’s resources and there is now a greater appreciation than ever before of the affect of man-made pollutants and wastes on the environment and their impact on climate change. The compounds studied by mass spectrometry include heavy metals, man-made pesticides, components of industrial waste and their by-products, disinfecting agents, explosives, and those excreted or obtained from naturally occurring algae, toxins and microorganisms. These substances are often present at trace levels within complex mixtures and can only be studied using a mass spectrometer.

9.1.1 Heavy Metals and Elemental Analysis Although trace concentrations of some metals such as iron and zinc obtained from certain foods are essential to our well-being, others such as lead and mercury have long been known to cause serious adverse effects to human health. Exposure to lead in paints, petrol and other industrial waste, for example, has long been associated with mental illness and even paralysis. Lead has four natural isotopes 204Pb,”‘Pb, 207Pb and 208Pbof which the latter three are produced by radioactive decay of isotopes of uranium and thallium. 9. I . 1.1 Thermal Ionisation Mass Spectrometry. Thermal ionisation mass spectrometry (TIMS) is a common technique applied to the 160

Mass Spectrometry in the Environmental and Surface Sciences

161

analysis of elements at low ng g-’ levels in environmental samples. In TIMS, a small volume of an aqueous sample solution containing between pg and ng of the element of interest is deposited onto a clean filament surface and evaporated to dryness. The filament, usually composed of a thin film of ruthenium, is heated to thermally evaporate the sample. A second like-filament that emits electrons to ionise the sample. Thermal ionisation is highly selective, with different elements ionised according to the filament temperature at any point in time. Isotopes are separated and detected simultaneously using a magnetic sector mass spectrometer equipped with a multiple ion collector. 9.1. I .2 Inductively Coupled Plasma Mass Spectrometry. Inductively coupled plasma mass spectrometry (ICP-MS) is also widely used for elemental analysis (Figure 9.1). The sample, usually in a liquid form, is introduced at approximately 1 ml min-’ via a nebuliser where it is converted into a fine aerosol with a gas (normally argon). The fine droplets of the aerosol are transported into the plasma torch via a sample injector where they are converted from a liquid aerosol to a solid and then to a gas. This sample gas is then “atomised” and ionised. Ionisation is achieved by electron impact using electrons from a high-voltage spark. This results in a high temperature (-7,000 K) ion plasma being emitted from the open end of the tube which reflects the elemental composition of the sample. Solid samples can be analysed directly by laser ablation ICP-MS. Here a portion of the sample is ablated by a laser in an inert atmosphere (usually of argon) under atmospheric pressure and then atomised and ionised in the plasma. Most commercial ICP mass spectrometers contain a single ion detector to detect elements to low part-per-trillion (ppt) levels. However, specialised magnetic sector ICP-MS instrumentation fitted with multiple detectors are also used for isotope ratio analysis (see also Section 9.2). The concentrations of heavy metals such as chromium (Cr) in soil and nebulizer I

Plasma torch

quadrupole MS

sample Ar

-

RFpower

Figure 9.1 Schematic representation of an inductively coupled plasma mass spectrometer (ICP-MS) ,featuring a quadrupole mass analyser

162

Chapter 9

water samples have been analysed by ICP-MS. Cr(V1)-containing compounds are found in the environment as a result of industrial pollution and fertilisers. Ion exchange chromatography has been coupled to ICP-MS to measure chromium levels in waters from industrial waste and sewage treatment plants. Both cationic Cr(II1) and anionic Cr(V1) species were detected below 0.5 pg/1 by monitoring 52Crusing TCP-MS. A disadvantage of ICP-MS is that there is a multitude of plasmaderived chemical reactions that can occur, resulting in molecular ions that can mask or diminish ions of the elements of interest. In order to improve detection limits and/or reduce the amount of sample that needs to be analysed, alternate low-pressure helium sources have been constructed with collision cells used to dissociate molecular ions formed within the plasma. 9.1.1.3 Isotope Dilution. Isotope dilution is a method used in conjunction with TIMS and ICP-MS to quantitate the elemental abundances of trace elements that contain two or more naturally occurring isotopes. It does so with unmatched sensitivity and accuracy and thus is in widespread use. The isotope dilution method involves spiking and blending the sample (S) of interest with a reference (R) material containing the same element at known concentration. The quantitation of trace elements in the samples is then determined by equation 9.1 where Q, is the element concentration of interest, QR is the concentration of the element in the reference material, R the ratio of isotopes in the reference, S the isotope ratio of the same isotopes in the sample, B the isotope ratio of the elements in the blend, m, the atomic mass of the lighter isotope for the element, and m,the atomic mass of the heavy isotope.

To minimise errors, the purities of samples S and R must be approximately the same and the mass spectrometric measurements recorded under the same conditions. When this is observed, measurements to within 0.1% of true values are possible. TIMS in conjunction with the isotope dilution method has been used in environmental research to determine heavy metals in the atmosphere. The presence of thallium in the atmosphere over Antarctica was measured at 0.2 pg m-3 using this approach.

9.1.2 Organic Pesticides G C and LC coupled mass spectrometric-based methods are preferred over other analytical approaches for the characterisation and

Muss Spectrometry in the Environmental and Surfuce Sciences

163

quantitation of organic pesticides. New pesticides are developed every year, in part to keep abreast of the resistance of insects and other pests to existing ones. For each new pesticide, a thorough evaluation of its fate in the environment is required including knowledge of its transport properties and degradation products. Pesticides can accumulate in soils, infect ground water and water supplies and be ingested by humans by way of crops and livestock. An accurate measure of the levels of pesticides in the environment is routinely needed in order to minimise their adverse effects on human health. Some modern pesticides are active as a single enantiomeric form such that regulatory authorities mandate that only this enantiomer may be administered where the inactive enantiomer has a low degradation rate. This necessitates that enantiomeric forms of a pesticide in a racemic mixture are resolved and independently quantified in environmental samples. As an illustration, soil samples from the south-east of Spain treated with the propionic acid-derived herbicides mecoprop and dichloroprop were characterised by HPLC using a chiral stationary phase and GCMS. Selected ion chromatograms from the GC-MS analysis of mecoprop (MCPP) (mlz 169) and dichlorprop (DCPP) (mlz 189) show that the methylated forms of R,S-MCPP and R,S-DCPP are clearly resolved in two soil samples (denoted a and b in Figure 9.2). Quantitation of the ion signals demonstrated detection levels of greater than 80% for both MCPP and DCPP from control soil-doped experiments. For each pesticide, the R-enantiomer was found to degrade faster than its inactive S-form.

9.2 ISOTOPE RATIO MASS SPECTROMETRY Beyond the characterisation and quantitation of compounds in the environment, isotope ratio mass spectrometry can be used to distinguish the source of such chemicals. The uptake of water, nitrogen and carbon in plants, nutrient transfer in aquatic ecosystems, and the rates of chemical and biochemical degradation processes can all be studied by mass spectrometry. Isotope Ratio Muss Spectrometry (IRMS) is performed using a specially constructed mass spectrometer designed to maximise ion beam stability and sensitivity at the expense of mass resolution. These instruments (Figure 9.3) usually feature an electron impact source with a gas inlet. As in a single magnetic sector mass spectrometer (described in Section 3.3.2), ions are accelerated down a flight tube between a magnet where their curved trajectories depend on the ion’s mlz ratio (equation

164

Clmpter 9

808

. 988

13.33

15.90

I

"

'

8

'

".I----.

116.66

.

"

' I

''.

1100 18.33

'.

'

I

'

"

lzoo 19.99

'

.' " I

c

YIY 4

(a)

1

I .

"

I

"

no0

14e4

21.1%

23.33

1500

Scan number/lime (min I

Figure 9.2 GC- M S chromatograms of methylated R- and S-MCPPplus R- and S-DCPP in a silty loam sample ( a ) without and ( b ) with addedpeat. In both cases, the total-ion chromatogrum and the single-ion chromatograms corresponding to mlz values of169 (spec@ for the enantiomers o j M C P P ) and 189 ( f o r those qf DCPP) are shown (Source: F. Sanchez-Rasero, M.B. Matallo, G. Dios, E. Romero and A. Pefia, Simultaneous determination and enantiomeric resolution of mecoprop and dichlorprop in soil samples by high-performance liquid chromatography and gas chromatography - mass spectrometry, J. Chromatography A , 1998, 799(1-2), 355-360, Figure 2)

3.22). Instead of scanning the magnetic field or accelerating voltage, these values are fixed for a particular measurement to transmit to the Faraday cup detector (Section 3.4.1) ions with a mlz range of just a few mass units. The mass limit of an isotope ratio mass spectrometer is typically 100. Slits on the instrument are large to maximise ion transmission. As a result, the ion signals typically have a flat top (Figure 9.4). Isotope ratios are measured based upon the area under the isotopic ion signals. The analytes are often simple gases such as H,, O,, N, and CO, or pyrolysis or combustion products formed by heating of the sample at

Muss Spectrometry in the Environmental and Surface Sciences

G a s inlet

165

--+

Figure 9.3 Schematic diagram of an isotope ratio mass spectrometer (Source: A. Barrie and S.J. Prosser, in Mass Spectrometry of Soils, T.W. Boutton and S.-I. Yamasaki (ed), Marcel Dekker, New York, 1996, Ch. 1, p. 9, Figure 1)

high temperatures. A gas chromatograph can be used to separate these combusted species prior to entry into the ion source. 9.3 PORTABLE MASS SPECTROMETERS There are many instances where it is desirable to analyse environmental compounds or organisms in the field. A variety of miniature, portable mass spectrometers have been used for this purpose. Small quadrupole ion trap and time-of-flight mass analysers are typically employed over heavier magnetic-based instruments. These truck or hand-portable mass spectrometers are used by government laboratories, academic researchers and by the military in applications as diverse as environmental monitoring, forensic science, oceanographic research, and the detection of nerve gas and other agents of warfare. Field portable GC-MS mass spectrometers feature a sensing device or sniffer with which to sample volatile agents in air. Submergeable mass spectrometers have also been developed for oceanographic research to

166

Chapter 9

28

29

m/z

Figure 9.4 Ion signals measured on u isotope ratio muss spectrometer typically have aflut top (Source: adapted from A. Barrie and S.J. Prosser, in Muss Spectrometry of Soils, T.W. Boutton and S.-I. Yamasaki (ed), Marcel Dekker, New York, 1996, Ch. 1, p. 10, Fig. 2)

monitor water quality, including the influences of tidal flows or the effects of dredging on water quality. These analytical measurements have particular challenges for a mass spectrometer including maintaining a vacuum in such an instrument underwater, providing a sufficient power source during the course of the experiments, transmitting the data to the shore, and sampling the analytes underwater. This latter challenge can be met by using membrane introduction mass spectrometry (MIMS) in which a sample is exposed to a semi-permeable membrane through which compounds are selectively transferred into the mass spectrometer. Components, sampled from such environments, can be analysed in concentrations down to the parts-per-trillion level. Despite their ban under the Geneva Protocol of 1925, chemical warfare agents have continued to be produced and in some cases inflicted on both military and civilian populations. Military personnel are particularly susceptible to exposure to chemical weapons. These include mustard gas and lewisite that cause blistering of the skin, diphenylcyanoarsine that leads to vomiting, tearing agents and the nerve agents sarin, tabun and so-called VX. Aircraft, ground-based vehicles or personnel can carry portable mass spectrometers to check for such agents prior to the deployment of large numbers of troops. The EI mass spectrum of mustard gas or l,l-thiobis(2-chloroethane)is shown in Figure 9.5. Mustard gas is a highly fat soluble compound and accumulates in tissues with a high fat content. Absorption of just a few milligrams of mustard gas within human tissue can be fatal.

Mass Spectrometry in the Environmental and Surface Sciences o

167

l

- CH2C1

Cl-CH2CH2-S-CH2CHZ-Cl 80

60

-1 M+.

; 120

40

80

160

2

K)

m/Z

Figure 9.5 EI mass spectrum ojmustard gas or I , 1-thiobis(2-chloroethane) (Source: RSC database)

9.4 CHEMISTRY OF THE EARTH’S IONOSPHERE The Earth’s atmosphere and ionosphere are rich in ion chemistry made possible due to the production of ions by electrical discharges and radiation. A significant understanding of the chemistries above the surface of the planet has been achieved by experiments that mimic such processes in the laboratory and by in situ measurements using rocket-borne mass spectrometers. Of particular interest is the effect of man-made pollutants on the Earth’s atmosphere. The Earth’s atmosphere is divided up by altitude according to a range of criteria where the ionosphere is described as the region where free electrons exist at some 60-1000 km above the Earth’s surface. At the highest altitudes, levels of solar radiation are the greatest but since there are few atoms or molecules present, few ions form. Below the ionosphere, electrons spontaneously react with gases to form both positive and negative ions. From the standpoint of ion chemistry, most can be classified as occurring at low altitudes below some 80 km, or at high altitudes or low pressures above this height. The lower atmosphere above 10-1 5 km is defined as the stratosphere and below that the troposphere where the major gases present are oxygen (21%) and nitrogen (78%) together with argon and varying amounts of water, carbon monoxide and dioxide, and nitrogen dioxide. In the upper atmosphere, solar radiation reaches the earth at a power of 1.37 kJ rnp2s-’. The photoionisation of an atom of oxygen by solar

168

Chupter 9

radiation in the ionosphere results in 0'' (Chapter 2, equation 2.1) that can subsequently react to form the molecular ions NO'' and O,+ (equations 9.2 and 9.3). 0''

+ N, + NO" + N'

(9.2)

+ 0'

(9.3)

O"+ 0, + 0;

Rates for these reactions have been measured in the laboratory within high pressure and flowing afterglow mass spectrometers to be approximately 1.2 x lo-', cm3 s-' and 2 x lo-'' cm3s-' respectively at 300 K. Cluster ions are also evident in the ionosphere formed by the weak association of water, nitrogen (equation 9.4) and carbon dioxide with ions in three-body associations. NO"

+ nH,O + N, S NO"

(H,O),(N,)

(9.4)

Cosmic rays are the primary ionisation source in the stratosphere and upper troposphere. One reaction of interest within the stratosphere is the recombination of molecular oxygen with oxygen atoms to form ozone. This gas reaches a peak density of a few parts per million at an altitude of 25 km above the Earth. The ozone layer protects the Earth from ultraviolet (UV) radiation that can penetrate the atmosphere by absorbing this radiation and dissociating back to its constituents. Negatively charged ions have been postulated to regulate ozone levels in the lower atmosphere as illustrated in equation 9.5 though this has been questioned based on a measured rate for this reaction. NO;

+ 0, + NO; + 20,

(9.5)

It has further been shown that most ions do not react with man-made pollutants such as trichlorofluorocarbon (CFCl,) and as such these pollutants cannot be depleted before they reach the stratosphere where they degrade ozone levels.

9.5 MASS SPECTROMETERS IN SPACE It might be argued that the best environment in which to operate a mass spectrometer is one in natural vacuum. Beyond the environment within and around our own planet, mass spectrometers have been employed in space to study the cosmos. Expeditions to Mars, investigations of interstellar dust and the chemical composition of the tails of comets have all involved mass spectrometers. Earlier mass spectrometry experiments were performed on rocks returned to Earth during the Apollo missions.

Mass Spectrometry in the Environmental and Surface Sciences

169

9.5.1 Apollo Missions Samples of the lunar surface returned to Earth as part of the six Apollo missions 11-17 (except the ill-fated Apollo 13 mission!) were subsequently examined by mass spectrometry. NASA scientists used secondary ion mass spectrometry (SIMS) (Section 3.2.5) to analyse the elemental composition of the moon rocks. This data has shown the moon to be enriched in the elements aluminium, uranium and thallium, in addition to ferric oxide, over that found on Earth. Samples obtained from the lunar highlands are rich in potassium, phosphorous and rare earth elements. The distribution of these elements on the lunar surface provides information about how the lunar crust was formed and has evolved over time. Samples of finely pulverised lunar surface material were also analysed for traces of organic compounds by GC-MS. Samples were volatilised directly, after extraction with benzene and methanol, or extraction and pre-treatment with hydrogen fluoride and chloride. A number of organic compounds were detected at concentrations of less than 1 ppm, but all could have been associated with contamination on Earth and therefore no evidence of life (as we know it) on the surface of the moon has been found to date.

9.5.2 Viking and Mars Express Missions

To preclude the possibility of contamination of returned specimens, the objectives of the Viking experiments were to sample the composition of the atmosphere at the surface of Mars and to identify any volatile organic and inorganic compounds on the surface at the landing site. These data were measured directly at the surface of Mars and transmitted electronically back to Earth. The Viking spacecraft of the 1975 mission transported a Mars lander fitted with a gas chromatograph mass spectrometer designed for isotope ratio experiments. Analysis of the atmosphere found the isotopic levels of oxygen to be within 10% of those values on Earth. A higher level of 15N(74%), however, was detected at the surface of Mars with an isotope ratio of 15N/14N of 0.0064 (without background correction). 38Arwas also detected above the Martian surface. Soil samples collected at the Martian surface were robotically loaded into a pyrolysis source at the lander site and heated to 500 "C in a series of three chambers. The volatiles were passed into the carrier gas stream of a gas chromatograph interfaced to a mass spectrometer operating over

170

Chapter 9

a mass range of 12-200. The data, together with operating parameters, were transmitted back to Earth from the Martian surface. Only water and carbon dioxide were detected in these measurements and there was no evidence or organic compounds above the detectable levels of a few parts-per-billion. The absence of such compounds at the lander site does not preclude that life as we know it could be found elsewhere on the planet surface or interior. It is also possible that traces of microorganisms in the samples did not release sufficient levels of organic material for detection or that such organic matter had been destroyed on the surface by solar radiation. An opportunity to re-evaluate the planet's surface was afforded by the European Mars Express mission of late 2003 in a British-led expedition. The Beagle 2 lander carried an onboard minature 90" sector mass spectrometer with a magnet weighing less than 1 kg. It was fitted with a dual inlet source so that light elemental samples and standards can be sequentially studied in high precision isotope ratio measurements. Regrettably, signals from Beagle 2 were lost leaving its fate unknown and mysteries of the Red planet firmly intact.

9.5.3 Composition of a Comet Unlike Mars, there is little doubt that organic compounds are ubiquitous within the nuclei of comets. There is considerable data to support the existence of molecular species within a comet. Data from ground-based spectroscopic measurements have been supplemented by both spectroscopic and mass spectrometric measurements in the Giotto and Vega missions to Halley's comet. Water makes up about 80% of the volatile content of the comet but hydrogen cyanide, carbon monoxide and dioxide, methanol, ammonia and formaldehyde have all been detected. Many of these molecules are likely to be ionised fragments of even larger parent molecules. The identity of these parent molecules, however, is not known since it is virtually impossible to observe the comet surface directly, even when a spacecraft is nearby. The dust from Halley's comet was also examined using mass spectrometers on board both the Giotto and Vega probes. Of primary interest was the detection of intermediate-sized organic compounds that gave rise to mass spectra with ion signals separated by repeating 14-1 6 mass units. These ions indicated that the molecules exhibit a linear polymeric structure interpreted to be a signature of hydrocarbons, with a repeating -(CHJn- structure. The Giotto and Vega spacecraft further revealed a substantial enrichment of 12Cover that observed on Earth indicating an interstellar source for some of the organic compounds.

Muss Spectrometry in the Environmental und Surface Sciences

171

9.6 APPLICATIONS OF SECONDARY ION MASS SPECTROMETRY TO MATERIALS SCIENCE Beyond moon rocks and interstellar particles, secondary ion mass spectrometry (SIMS) (Section 3.2.5) is in widespread use for the analysis of solid surfaces and materials, including thin films and semiconductors. SIMS is used to detect atomic ions as well as molecular ions, the latter often detected as clusters. Detection limits of the order of 10l2and 10l6 atoms per cubic centimetre. Mass interferences (peaks from different molecules or atoms that share a common mlz value) are a common feature of SIMS experiments, and it is necessary to anticipate them in advance such possibilities in the design of an experiment.

9.6.1 Depth Profiling Since the primary ion beam can be focused to a diameter of less than one pm, SIMS provides a means with which to characterise a surface with high resolution. Where continuous sputtering of the surface is performed, an analysis of the material as a function of depth (a depth profile) can be produced. Typical surface depths are of the order of 1 nm. This analysis is useful in industrial applications in order to study the quality of manufactured coatings or the processes used to construct them. A SIMS depth profile of a stainless steel surface coated with layer of titanium is shown in Figure 9.6. The figure shows that the titanium coating has a depth of 1.4 pm. One application of SIMS is ion microscopy. Here the primary beam is focused on the sample over an area of approximately 10 pm. Secondary ions released from the surface are passed into an electrostatic mirror in which they are energy-focused and reflected back to the mass analyser and onto an image converter. The image converter translates the spatial distribution of the atoms on the surface onto a fluorescent screen for visualisation. In combination with a depth profile, ion microscopy enables three-dimensional maps of a material over a diameter of 250 pm to be constructed with resolutions of the order of 1 pm.Larger surface areas can also be studied though usually at reduced resolution. 9.6.2 Analysis of Impurities The characterisation of surfaces containing aluminium, silicon, tungsten, gallium and titanium has been accomplished using SIMS including the identification of impurities such as oxygen. The strength

172

Chapter 9

Figure 9.6 SIMS depth projile of a stainless steel surface coated with layer c?f'titanium. The titanium has a depth of 1.4,urn

and mechanical properties of metals and polymers can be substantially altered by the presence of inert impurities or gas pockets. SIMS has found particular application in the semiconductor industry since impurities or doped elements incorporated into the material can either reduce or enhance the electrical properties of the material. Semiconductors are usually constructed of single crystals of silicon and gallium within which undesired impurities must be kept well below the 1YOlevel. SIMS analysis can determine the distribution of trace levels of impurities (down to 10 pg per gram of material) in high purity materials in three-dimensions. In the case of borosilicate glass, it can reveal impurities such as lithium and sodium at particular depths that can be responsible for weaknesses and subsequent fractures. It can also be exploited to examine glass coatings used for optics in scientific and industrial applications.

9.6.3 Reaction Catalysts

SIMS has been further applied to the study of reaction catalysts in terms of their molecular structure and that of their clusters. Transition metal complexes are one type of catalyst that can promote chemical transformations without being consumed during a reaction. The structures

Mass Spectrometry in the Environvlzental and Surface Sciences

173

13co 16

I4

0

? z

40

l2 10

‘U 8

u1

0 Q

IU

6 4

(-

) calculated

2

O

L

0

’

I

5

10

15

20

I

I

I

25

30

35

1

I

I

I

40

45

50

55

TIME (min)

Figure 9.7 Conversion ojcarbon monoxide into methane above a nickel catalystJollowed by SIMS using a ”C tracer (Source: M. Otarod, S. Ozawa, F. Yin, M. Chew, H.Y. Cheh and J. Happel, Multiple isotopic tracing of methanation over nickel catalysts, J Catalysis, 1983,84, 156-1 59)

of metal complexes such as nickel oxide have been studied on polymeric supports and oxide coatings. The poisoning of such catalysts has also been investigated by mass spectrometry. Nickel oxides act as catalysts to promote the conversion of carbon monoxide and molecular hydrogen into methane above their surface which can be followed by SIMS (Figure 9.7). When hydrogen disulfide is used to “poison” the catalyst, a reduction in the formation of methane can be followed as a function of the proportion of the nickel oxide surface covered with sulphur.

FURTHER READING J.R. de Laeter, Applications of Inorganic Mass Spectrometry, WileyInterscience, New York, 2001. A. Montasser, Inductively Coupled Plasma Mass Spectrometry, Wiley-VCH, Berlin, 1998. J.D. Rosen (ed) Applications of New Mass Spectrometry Techniques in Pesticide Chemistry, John Wiley & Sons, New York, 1987. I.T. Platzner, Modern Isotope Ratio Mass Spectrometry, John Wiley & Sons, New York, 1987. E.R. Badman and R.G. Cooks, Miniature mass analyzers, J Muss Spectrom., 2000,35(6), 659-671.

174

Chapter 9

T. Kotiaho, F.R. Lauritsen, T.K. Choudhury, R.G. Cooks and G.T. Tsao, Membrane introduction mass spectrometry, Anal. Chem., 1991, 63(18), 875A-883A. R.C. Murphy, G. Preti, M.M. Nafissi-Varchei and K. Biemann, Search for organic material in lunar fines by mass spectrometry, Science, 1970,167(3918), 755-7. K. Biemann, J. Oro, P. Toulmin, L.E. Orgel, A.O. Nier, D.M. Anderson, P.G. Simmonds, D. Flory and AS? Diaz, Search for organic and volatile inorganic compounds in two surface samples from the Chryse Planitia region of Mars, Science, 1976, 194(4260), 72-76. T. Owen, K. Biemann, D.R. Rushneck, J.E. Biller, D.W. Howarth and A.L. LaFleur, The atmosphere of Mars: detection of krypton and xenon, Science, 1976,194(4271), 1293-1295. A.O. Nier, Mass spectrometry in planetary research, hit. J Mass Spectrom. Ion Proc., 1985, 66, 55-73. M.R. Sims, C.T. Pillinger, I.P. Wright, J. Dowson, S. Whitehead, A. Wells, J.E. Spragg, G. Fraser, L. Richter, H. Hamacher, A. Johnstone, N.P. Meredith, C. de la Nougerede, B. Hancock, R. Turner, S. Peskett, A. Brack, J. Hobbs, M. Newns, A. Senior, M. Humphries, H.U. Keller, N. Thomas, J.S. Lingard and T.C. Ng, Beagle 2: a proposed exobiology lander for ESA’s 2003 Mars Express mission, Adv. Space Res., 1999,23(11), 1925-1928. A. Benninghoven, in Secondary ion mass spectrometry: basic concepts, instrumental aspects, applications, and trends, A. Benninghoven, F.G.R. Denauer, H.W. Werner (ed), John Wiley & Sons, New York, 1987. H. Oechsner, Inorganic mass spectrometry for surface and thin film analysis, Anal. Chim. Acta, 1993,283, 131-138. J. S. Becker and H.-J. Dietze, Inorganic trace analysis mass spectrometry, Spectrochimica Acta B, 1998,53, 1475-1 506. J.G. Holland and S.D. Tanner (ed), Plasma Source Mass Spectrometry: The New Millennium, Royal Society of Chemistry, Cambridge, UK.

CHAPTER 10

Accelerator Mass Spectrometry 10.1 INTRODUCTION Accelerator Mass Spectrometry (AMS) is an analytical technique that uses an ion accelerator as an ultrasensitive mass spectrometer to ultimately count individual atoms. AMS was first introduced in 1977 and constitutes a highly sensitive method for detecting very low concentrations of long-lived radioisotopes or stable isotopes in a wide range of samples. AMS separates rare radioisotopes from stable ones and measures their relative ratio with high sensitivity and precision. It is commonly used in radiocarbon dating experiments, where carbon-based materials are converted to graphite, and the amount of 14Cthey contain is measured. This provides a measure of the age of the item based on the half-life of the 14C isotope of 5568 years. Meteorites from space, air trapped in Antarctic ice, and the Turin shroud are some of the sources of samples to which AMS has been applied. Other applications include studying radiolabelled tracers in biological systems. In the AMS technique, the element of interest is chemically separated from the original sample and loaded as a target in the sputter ion source of the tandem accelerator (Figure 10.1). Samples are pulverised, treated with acid and alkali and freeze-dried. In the case of carbonbased compounds, the sample is converted to either graphite or carbon dioxide. After ionisation of the samples and the separation of ions using a magnet, negative ions containing the radioisotope of interest are accelerated through a potential of several million volts (MV). Negative ions are used to distinguish 14C from I4N since the latter does not form a negative ion. A gas such as sulphur hexafluoride is added to the accelerator to dissociate all molecular ions to an atomic form. At the end of this first acceleration stage these ions pass through a stripper. A stripper consists of a thin carbon foil or gas that strips electrons from ions and destroys any molecular isobars. In the case of carbon, any 12CH; and 13CH+ions are fragmented to leave only 14C+ions with a mlz of 14. These positive ions are further accelerated to energies of up 175

176

Chapter 10

to several tens of MeV in the second stage of the tandem accelerator. Acceleration of the ions to high energies enables ions to be uniquely identified based on their total energy. Using a magnetic and electrostatic mass analyser, the ions are focused into a Faraday cup for ion detection. For carbon samples, the ratio of l4C/I3Cand 14C/'2Cis measured and compared to measurements made for standards of known ratios. It is possible to measure isotopic ratios down to 1:10-15,or low attomole levels.

Figure 10.1 Schematic representation of a tandem accelerator muss spectrometer featuring dual ion sources, preaccelerator and tandem accelerator ( A C C ) (Source: C. Tuniz, J.R. Bird, D. Fink and G.F. Herzog, Accelerator Mass Spectometry: Ultrasensitive Analysis jor Global Science, CRC Press, Boca Raton, 1998, Ch. 1, p. 42, Fig. 3.1)

AMS is unaffected by almost all background effects that limit conventional mass spectrometry measurements. Thus AMS is five to ten orders of magnitude more sensitive than a conventional mass spectrometry experiment. The amount of sample required for accelerator mass spectrometry is far less (typically a few mg) than that required for beta-particle decay counting, with around 1% of all the 14Cin a sample measured.

10.2 ION SOURCES

A typical AMS source consists of a heated reservoir of caesium powder, an ioniser that produces a Cs' beam focused at the sample, and an extraction electrode to accelerate and focus secondary negative ions

Accelerator Muss Spectrometry

177

from the sample (Figure 10.2). Solid samples are deposited to a diameter of 1-2 mm in the centre of a cooled metal plate (usually cast of copper or aluminium) which acts as the cathode. The caesium ions are accelerated toward the cathode and on impact sputter or release particles from the surface. Such sputter sources efficiently produce negative ions for many elements and molecules depending on their electron affinity. The negative ions are injected into the accelerator where all molecular ions are dissociated and positive ions are formed.

*

c-

Cs Vapor Figure 10.2 Schematic representation of a typical accelerator mass spectrometry ion source (Source: adapted from C. Tuniz, J.R. Bird, D. Fink and G.F. Herzog, Accelerator Mass Spectometry: Ultrasensitive Analysis for Global Science, CRC Press, Boca Raton, 1998, Ch. 1, p. 43 - part (b), Fig. 3.2)

10.3 PERFORMANCE AND LIMITATIONS OF RADIOCARBON DATING Many radioisotopes including 3H, I4C, 26Al,32Si,39Arand *lKr are produced in the atmosphere by cosmic rays through nuclear reactions. 14Cis produced by reactions between stable 14N nuclei and neutrons in the upper atmosphere and is subsequently converted to carbon dioxide. This carbon dioxide is assimilated into plants, and through their consumption into animals. An equilibrium is maintained in the Earth’s atmosphere, hydrosphere and biosphere through the continuous production of atmospheric 14C.The stable isotopes of carbon (12C and I3C) constitute the majority of carbon on Earth (98.9% and 1.1% respectively). The level of I4C on Earth, in contrast, is extremely low and has been measured to be about of all carbon. The measurement of 14Cis used in numerous applications of which radiocarbon dating is the best known. Radiocarbon dating involves measuring the I4C in biological specimens or archaeological relics to

178

Chapter 10

calculate their age. All living organisms or protected archaeological relics contain nearly the same proportion of radioactive carbon at the time of death or burial. The level of 14C subsequently decreases by radioactive decay with a half-life of 5,568 years. By measuring the residual levels of 14Cin a sample, the age of the source material can be estimated. The precision of radiocarbon dating measurements depends on a number of factors including the amounts of material available for analysis, contamination of the sample, reservoir effects, and variations in 14Cproduction. For specimens less than 5000 years of age, a minimum of about 50 pg of sample material is required for analysis. Optimal sample levels are of the order of a few mg to as much as a gram. At the lower levels, errors of the order of 1% in the age in years are typical. Contamination is a major source of errors in AMS measurements particularly in older specimens. Bone carbonate, for example, is prone to the exchange of carbon from the environment particularly when buried in carbon-rich soils. The 14Ccontent in carbon dioxide trapped in polar ice cores can be measured providing it can be separated from that induced by radiation on the ice surface prior to it becoming buried. The so-called reservoir effect occurs when samples derive carbon not from the Earth’s equilibrium but from local environments. These effects are apparent in plants found near volcanoes that release I4C-depleted carbon dioxide and in deep sea aquatic systems.

10.4 APPLICATIONS OF RADIOCARBON DATING IN ARCHAEOLOGY AND COSMOLOGY The shroud of Turin, an ancient cloth that many Christians believe was used to wrap Christ’s body following his crucifixion, represents one of the most controversial radiocarbon dating experiments performed to date. The analysis was performed simultaneously at AMS facilities in Europe and the United States in 1988 from samples cut from the shroud in Turin, Italy. All three laboratories subdivided the samples, and subjected the pieces to different mechanical and chemical cleaning procedures to remove contaminants. The samples were further analysed microscopically to identify and remove any foreign material. All laboratories combusted the textile segments with copper and then converted the resulting C02 into graphite targets. Three to five separate measurements were made at each laboratory. The three laboratories in Arizona, Oxford and Zurich reported the age of the shroud at 641 k 31, 750 f 30 and 676 f 24 years respectively, far younger than is possible if the fabric had been used to wrap the body of Jesus Christ (Figure 10.3).

Accelerator Mass Spectrometry

179

1

2 0

5 c n 3 0

tM

* rti

* 4

-

H3-i 8 I

1

1000 800 600 Radiocarbon age (yr BPI

2000 1900

Figure 10.3 Mean radiocarbon dates, with a ?I standard deviation of the Shroud of’ Turin and control samples, as supplied by three laboratories ( A , Arizona; 0, Oxford; Z , Zurich). The age of the shroud is obtained as AD 1260-1390 with at least 95% confidence (Source: P.E. Damon, D.J. Donahue, B.H. Gore, A.L. Hatheway, A.J.T. Jull, T.W. Linick, P.J. Sercel, L.J. Toolin, C.R. Bronk, E.T. Hall, R.E.M. Hedges, R. Housley, I.L. Law, C. Perry, G. Bonani, S. Trumbore, W. Woelfli, J.C. Ambers, S.G.E. Bowman, M.N. Leese and M.S. The, Radiocarbon Dating of the Shroud of Turin, Nature, 1989,337(6208), 61 1-61 5)

Radiocarbon dating measurements can also investigate when early civilisations first occupied land or cultivated crops. Controversy still surrounds the initial occupation of land in Europe, the Americas and Australia. The oldest known human occupation of territories in Australia has been dated at some 40,000 years ago, far before European settlement. Rock art of the indigenous Australian aboriginals uncovered in a series of excavations in far north Queensland has been dated at 26,000 years. Accelerator mass spectrometry has also been used to determine the age of maize cobs in Oaxaco in Mexico. Radiocarbon dating recorded the age of the maize, an ancient corn, at over 6,200 years old, making them the oldest known agricultural crops in the Americas. The study of extraterrestrial materials from meteorite craters and rock are also of interest to scientists in order to establish that such materials are indeed extraterrestrial. Radiocarbon dating measurements can also be used to determine when such a meteorite collided with Earth and also its exposure to cosmic rays prior to collision in order to suggest its origin. The age of most specimens is evaluated based on I4C and 41Ca levels, the latter with a half-life of 130,000 years. Measurements of

180

Chapter I0

41Cahave the further advantage that its rate of productiori on Earth varies to a lesser degree than other isotopes.

10.5 BIOMEDICAL APPLICATIONS 14

C is used as a radioactive tracer in nuclear medicine both in medical research and for the testing of new pharmaceuticals on volunteers to follow metabolic processes and abnormalities. One method of following the metabolism of a 14C-labelled drug compound, that has been ingested and metabolised, is to collect and analyse the end-product carbon dioxide exhaled. Clinically useful information is usually obtained from carbon dioxide exhaled a few hours after the administration of the drug, even if the time required for its complete metabolism is much longer. This approach has been used to study the long-term retention of 14C-labelled triolein in fat by analysis of the patients' breath. There are, however, considerable errors in the estimates of the absorbed doses of 14C-labelledpharmaceuticals largely due to the long half-life of 14C. The metabolism of calcium is also of interest in the study of bone diseases such as osteoporosis. An imbalance between the excretion and absorption of calcium from bone is responsible for osteoporosis and can only be partly counteracted by diets rich in calcium due to its absorption through the intestines at approximately 30%. The calcium isotopes 4'Ca and 47Caare usually employed as tracers since their half-lives are far shorter than 41Ca(4.5 and 165 days respectively). In a bone absorption study in menopausal women, volunteers ingested 125 ng of radioactive calcium and their level of uptake was monitored by AMS measurements on their urine. The study's findings revealed a short-term rapid loss of calcium (by three orders of magnitude in 100 days) followed by a period of little loss (900 days). Studies of the uptake of 26A1from water sources have also been followed in rat brain by AMS in order to quantify the levels of environmental aluminium that enter the blood stream. The rate of passage of 26Alfrom the blood to the brain is of interest in the study of Alzheimer's disease and other neurological disorders. Other radioisotopes of importance in biomedical studies include due to the abundance of these elements in insecticides and 36Cland 1291 pesticides. Thus the levels of these compounds absorbed by humans can be measured as a function of exposure. These studies demonstrate the far-reaching applications of accelerator mass spectrometry. They further illustrate the scope of experimentation and discovery possible a century after the founding of mass spectrometry.

Accelerator Mass Spectrometry

181

FURTHER READING C. Tuniz, J.R. Bird, D. Fink and G.F. Herzog, Accelerator Mass Spectrometry: Ultrasensitive Analysis for Global Science, Boca Raton, Florida, CRC Press, 1998. H.E. Gove, From Hiroshima to the Iceman: The Development and Applications of Accelerator Mass Spectrometry, Institute of Physics, USA, 1998.

APPENDIX 1

Abbreviations used in Mass Spectrometry ADC AE AMS amu AP APCI B B/E

B21E CA CAD CE CF-FAB CI CID CR Da DE E EA ECD EE EI EM EST

analog-to-digital converter appearance energy accelerator mass spectrometry atomic mass unit, now u appearance potential atmospheric-pressure chemical ionisation magnetic sector mass analyser; magnetic field strength (in italics) magnetic sector linked scan constant for fragment ions magnetic sector linked scan constant for precursor or parent ions collisional activation collisionally activated dissociation capillary electrophoresis continuous-flow fast atom (or ion) bombardment chemical ionisation collision-induced dissociation charge reversal Dalton delayed extraction, also time-lag focusing (TLF) electric sector mass analyser; electric sector voltage electron affinity electron-capture dissociation even-electron ion electron ionisation; electron impact electron multiplier electrospray ionisation 182

Appendix 1

183

ESI-MS or ESIMS e

electrospray ionisation mass spectrum, spectrometry or spectrometer charge of an electron; one electron has a charge of 1.602 x Coulombs (C) electron volt flowing afterglow fast-atom bombardment field desorption field free region Fourier-transform ion cyclotron resonance, also FT-MS full-width at half maximum gas chromatography or chromatograph gas chromatography-mass spectrometry inductively coupled plasma mass spectrometry ion cyclotron resonance ionisation energy ion mobility spectrometry ion trap, also quadrupole ion trap (QIT) kilodalton kinetic energy release liquid chromatography mass spectrometry or spectrometer laser desorption ionisation (also LD) liquid secondary ion mass spectrometry matrix-assisted laser desorption ionisation massive cluster impact microchannel plate metastable ion mass-analysed ion kinetic energy spectrometry membrane introduction mass spectrometry multiple-reaction monitoring mass spectrometer, spectrometry or spectrum tandem mass spectrometry, spectrometer or spectrum; also MS2 molecular weight mass-to-charge ratio neutral molecule nanolitre flow electrospray ionisation neutralisationlreionisation orthogonal acceleration odd-electron ion proton affinity

eV FA FAB FD FFR FT-ICR FWHM GC GC-MS ICP-MS ICR IE IMS IT kDa KER LC-MS

LDI LSIMS MALDI MCI MCP MI MIKES MIMS MRM MS MSlMS MW mlz N nanospray NR oa OE PA

184

PAD PD PID PSD

Q 9 QET QIT RTOF rf SELDI SIC

SID SIFT SIM SIMS SRM t TIC TLF TOF U

V

z

Appendix I

post acceleration detector plasma desorption photon-induced dissociation post-source decay quadrupole mass filter quadrupole mass filter operating in rf-only mode quasi equilibrium theory quadrupole ion trap, also IT reflecting time-of-flight radio frequency surface-enhanced laser desorption ionisation selected ion chromatogram, also extracted ion chromatogram surface-induced dissociation selected-ion flow tube selected-ion monitoring secondary ion mass spectrometry selected reaction monitoring ion flight time, time total ion current time-lag focusing time-of-flight unit of mass volt, voltage charge of an ion; an integer multiple of e

APPENDIX 2

Isotope Masses and Abundances

I I

Isotope

I

'H

I

Nominal mass 1

1 1

Muss

1.007825032

I

Relative abundance

1

99.985(1)

*H;D

2

2.014101778

0.015( 1)

'H; T

3

3 .O 16049268

~0.0001

4He

4

4.002603250

-100

%i

6

6.0151223

7.5(2)

7Li

7

7 .O 160041

92.5(2)

9Be

9

9.0121822

-100

I0B

10

10.012937 1

19.9(2) ~

~

~

_

_

_ ~

~

IIB

11

11.0093055

80.1(2)

I2C

12

12.000000000

98.93(8)

?C

13

13.003354838

I .07(8)

I

I

1

15.00010897 15.994914622

1

I

_

0.368(7) 99.757(16)

170

17

16.9991315

0.038(1)

l8O

18

17.999160

0.205( 14)

I9F

19

18.9984032

-100

1 20Ne

I

2o

I

19.992440176

185

I

90.48(3)

_

_

_

186

Appendix 2

Isotope

Nominal mass 21

2'Ne

I

22Ne

Mass

I

22

I

Relative abundance

20.99384674

0.27( 1)

21.9913855

9.25(3)

23Na

23

22.9897697

-100

24Mg

24

23.9850419

78.99(4)

"Mg

25

24.9858370

lO.OO(1)

I 26Mg

I

26

I

25.9825930

1

11.01(3)

2 7 ~ 1

27

26.98 15384

-100

28Si

28

27.97692653

92.22(2)

29Si

29

28.97649472

4.69( 1)

'OSi

30

29.97377022

3.09(1)

3lp

31

30.9737615

-100

32s

32

3 1.9720707

94.93(3 1)

3 3 s

33

32.9714585

0.76(2)

3 4 s

34

33.9678668

4.29(28)

I 36s

I

36

I

35.9670809

I

0.02( 1)

35c1

35

34.96885271

75.78(4)

3 7 c 1

37

36.96590260

24.22(4)

"Ar

36

35.9675463

0.3365(30)

38Ar

38

37.9627322

0.0632(5)

40Ar

40

39.962383123

99.6003(30)

39K

39

38.9637069

93.2581(44)

K

40

39.9639987

0.01 17(1)

4'K

41

40.9618260

6.7302(44)

40Ca

40

39.9625912

96.941(156)

42Ca

42

41.9586183

0.647(23)

40

I 43ca

I

43

I

42.9587668

I

0.135(10)

187

Appendix 2 Isotope

Nominal mass

Mass

Relative abundance

44Ca

44

43.955481

2.086(110)

46Ca

46

45.953693

0.004(3)

4RCa

48

47.952533

0.187(21)

45sc

45

44.955910

-100

45.952630

8 . 2 33)

46.951764

7.44(2)

I

46Ti

46

I

47Ti

47

I I

48Ti

48

47.947947

73.72(3)

49Ti

49

48.947871

5.41(2)

50Ti

50

49.944792

5.18(2)

50v 5'v

50

49.947 163

0.250(4)

51

50.943964

99.750(4)

50Cr

50

49.946050

4.345(13)

52Cr

52

5 1.940512

I

53Cr

53

I

52.940653

54Cr

54

53.938885

55Mn

55

54.938049

I

54Fe

I

56Fe

I

5

y

~

~

I

I

-100

I

53.939615

I

56

I

55.934942

I

I

56.935398

9.501(17) 2.365(7)

54

57Fe "Fe

83.789(18) ~~~~

I

~~

5.845(35) 91.754(36) 2.119(10)

58

I

57.933280

I

0.282(4)

59

I

58.933200

I

-100

58Ni

58

57.935348

68.0769(89)

60Ni

60

59.930790

26.223l(77)

61Ni

61

60.931060

1.1399(6)

62Ni

9"

I

62 64

I

61.928348 63.927969

I

3.6345(17) 0.9256(9)

188

Appendix 2

Do tope 63cu

"cu

Mass

Nominal muss

I

63

I

65

I

62.929601

I

64.927794

Relative abundance

1

69.17(3)

I

30.83(3)

64Zn

64

63.929146

48.63(60)

"Zn

66

65.926036

27.90(27)

(j7Zn

67

66.927131

4.10( 13)

6sZn

68

67.924847

18.75(51)

70Zn

70

69.925325

0.62(3)

69Ga

69

68.925581

60.108(9)

7'Ga

71

70.924707

39.892(9)

70Ge

70

69.924250

20.84(87)

72Ge

72

7 1.922076

27.54(34)

13Ge

73

72.923460

7.73(5)

74Ge

74

73.921 178

36.28(73)

76Ge

76

75.92 1403

7.61(38)

7SAs

75

74.92 1597

-100

74Se

I

74

1

73.922477

I

0.89(4)

76Se

76

75.919214

9.37(29)

77Se

77

76.919915

7.63( 16)

78Se

I

78

I

77.917310

I

23.77(28)

*OSe

80

79.9 16522

49.6 1(4 1)

82Se

82

8 1.916700

8.73(22)

79Br

I

79

I

78.918338

I

50.69(7)

*'Br

I

81

1

80.916291

I

49.31(7)

I

78

78Kr *'Kr **Kr

I

I I

80 82

I

77.92039

0.35( 1)

1

79.916379

I

2.28(6)

I

81.913485

I

11.58(14)

189

Appendix 2

I

Isotope

Nominalmass

I

Muss

I Relative abundance

"Kr

83

82.914137

1 1.49(6)

84Kr

84

83.91 1508

57.00(4)

I

86Kr "Rb

I

s7Rb

I

86

85 87

1

I I

85.910615 84.911792 86.909186

I

17.30(22)

I I

72.17(2) 27.83(2) ~~

s4Sr

84

83.913426

0.56(1)

%r

86

85.909265

9.86(1)

87Sr

87

86.908882

7.00( I)

"Sr s9Y

90Zr

I

88

I I

89 90

1

I

I

87.905617

I

88.905849 89.904702

~

82.58(1) -100

I

51.45(40)

"Zr

91

90.905643

11.22(5)

92Zr

92

91.905039

17.15(8)

I

94Zr

94

I

93.906314

1

17.38(28)

96Zr

96

95.908275

2.80(9)

93Nb

93

92.906376

-100

92

9 1.906810

14.84(35 )

94

93.905087

9.25( 12)

9

2

~

~

9 4 ~ 0 9

5

~

~

95

94.905841

15.92(13)

9

6

~

~

96

95.904678

16.68(2)

9

7

~

~

I

97

9

8

~

~

98

97.905407

24.13(31 )

looMo

100

99.90748

9.63(23)

96R~

96

95.90760

5.52(20)

98Ru

98

97.90529

1.88(9)

99Ru

99

98.905939

12.74(26)

1

96.906020

I

9.55( 8)

190 Isotope

looRu "'Ru

Appendix 2

r I I

Nominal mass 100 101

I 1 I

Muss 99.904219

I I

Relative ahundunce

12.60(19)

100.905582

17.05(7)

Io2Ru

102

101.904349

31.57(31)

Io4Ru

104

103.905430

18.66(44)

Io3Rh

103

102.905504

-100

lo2Pd

102

101.905607

1.02(1)

IMPd

104

103.904034

11.14(8)

lo5Pd

I

105

I

104.905083

I

22.33(8)

Io6Pd

106

105.903484

27.33(3)

losPd

108

107.903895

26.46(9)

109.905153

11.72(9)

'loPd

I

110

I

Io7Ag

107

106.905093

51.839(8)

Io9Ag

109

108.904756

48.16 1(8)

Io6Cd

I

106

I

105.90646

I

1.25(6)

Io8Cd

108

107.90418

0.89(3)

"'Cd

110

109.903006

12.49(18)

"'Cd 'I2Cd

I I

111

112

1 I

110.904182 111.902758

1 I

12.80(12) 24.13(21)

3Cd

113

112.904401

12.22(12)

'I4Cd

114

113.903359

28.73(42)

'I6Cd

I

116

1

115.904756

I

7.49(18)

1131~

I

113

I

112.904062

I

4.29(5)

"51n

115

114.903879

95.7 1( 5 )

Il2Sn

112

111.904822

0.97( 1)

Il4Sn

114

113.902783

O.65( 1)

"'Sn

I

115

I

114.903347

I

0.34( 1)

Appendix 2

191 Nominul mass

Isotope

1 1

116 117

I 1

115.901745

14.45(9)

116.902955

7.68(7)

'I8Sn

118

117.901608

24.22(9)

19Sn

119

118.903311

8.59(4)

I2'Sn

120

119.902199

32.59(9)

122Sn

122

121.903441

4.63(3)

'24Sn

124

123.905275

5.79(5j

I2lSb

121

120.903822

57.21(5)

123sb

123

122.904216

42.79(5)

2oTe

120

119.90403

0.09(1)

122Te

122

121.903056

2.55(12)

123

122.904271

0.89(3)

124Te

124

123.902819

4.74( 14)

125Te

I25

124.904424

7.07(15)

1 2 3 ~ ~ ~

I 126Te

I

Relative abundance

Mass

~~

I

126

I

125.903305

I

18.84(25)

128Te

128

127.904462

3 1.74(8)

'OTe

130

129.906223

34.08(62)

1271

1

127

I

126.904468

I

-100

124Xe

124

123.905895

0.09(1)

I2%e

126

125.90427

0.09( 1j

I 128Xe

1

128

I

127.903531

I

I

1.92(3)

129Xe

129

128.904780

26.44(24)

I3OXe

130

129.903509

4.08(2)

I3'Xe

131

130.905083

21.18(3)

1 32xe

132

13 1.90415 5

26.89(6)

'34Xe

134

133.905395

10.44(10)

192

Appendix 2

Isotope

Nominal mass

Mass

1 36xe

136

135.90722

"CS

I

1 3 0 ~ ~

I

I I I

133 130 132 134

I

I I 1

132.905447

Relative abundance

8.87(16)

I

-100

129.90631

0.106(1)

131.905056

O.lOl(1)

133.904504

2.417(18)

~

1 3 5 ~ 3 ~

135

134.905684

6.592(12)

136Ba

136

135.904571

7.854(24)

1 3 7 ~ ~

I3*Ba 1 3 8 ~ ~

I I I

137 138 138

I I 1

136.905822 137.905242 137.907107

I I I

11.232(24) 71.698(42) 0.090(1)

"La

139

138.906349

99.910(1)

I3Te

136

135.90714

0.19( 1)

I 3sce

138

137.90599

0.2541)

I4'Ce

140

139.905435

88.48(10)

142Ce

142

141.909241

11.08(10)

I4'Pr

141

140.907648

-100

14*Nd

142

141.907719

27.13(12)

143Nd

143

142.909810

12.I8(6)

144Nd

144

143.910083

23.80( 12)

I4'Nd

145

144.912569

8.30(6)

1

146Nd

I

146

I

145.913113

I

17.19(9)

I4'Nd

148

147.916889

5.76(3)

"'Nd

150

149.920887

5.64(3)

'44Sm

144

143.911996

3.1(1)

'47Sm I4'Sm

I

147 148

I

146.914894 147.914818

I

15.0(2) I1.3(1)

Appendix 2

193

Is0t ope

Nominal mass

Mass

149Sm

149

148.917180

13.8(1)

lSOSrn

150

149.917272

7.4(1)

IS2Srn

152

151.919729

26.7(2)

IS4Srn

154

153.922206

22.7(2)

IslEu

151

150.919846

47.8( I . 5)

1 "*Gd

I

152

I

151.919789

1 IS4Gd

1

154

1

153.920862

Relative abundance

I

I

0.20(1)

1

2.18(3)

1

lssGd

155

154.922619

14.80(5 )

Is6Gd

156

155.922120

20.47(4)

Is7Gd

157

156.923957

15.65(3)

"'Gd

158

157.92410I

24.84( 12)

I6'Gd

160

159.927051

21.86(4)

IS9Td

159

158.925343

-100

155.92428

0.06(1)

157.924405

O.lO(1)

I 1 1

Ii6Dy F D y

[ '6'Dy

156 158 160

F D y - 61)

1

F D y

162

I I 1

I 1

159.925194

1

2.34(6)

160.926930

I

18.9(2)

161.926795

I

25.5(2)

163Dy

163

162.928728

24.9(2)

164Dy

164

163.929171

28.2(2)

165

164.930319

-100

162Er

162

161.928775

0.14( 1)

'"Er

164

163.929197

1.61(2)

166Er

166

165.930290

33.6(2)

1

6

P E r

5

~

~

1

167

1

1

1

166.932046

1

22.95(15)

I I

I

194

Appendix 2

Isotope

Nominal mass

Mass

168Er

168

167.932368

26.8(2)

I7'Er

170

169.935461

14.9(2)

169Tm

169

168.934211

-100

168Y b

168

167.933895

0.13(1)

17'Yb

170

169.934759

3.05(6)

"'Yb

171

170.936323

14.3(2)

17*Yb

172

171.936378

21.9(3)

173Yb

173

172.938207

16.12(21)

174Yb

174

173.938858

31.8(4)

176Yb

176

175.942569

12.7(2)

175Lu

175

174.940768

97.4 1(2)

I 176Lu

I

176

I

175.942683

Relative abundance

I

2.59(2)

174Hf

174

173.940042

0.162(3)

176Hf

176

175.941403

5.206(5)

177Hf

177

176.943220

18.606(4)

I7'Hf

178

177.943698

27.297(4)

179Hf

179

178.945815

13.629(6)

180

179.946549

35.100(7)

' *OTa

180

179.947466

0.0 12(2)

1 8 1 ~ ~

181

180.947996

99.988(2)

rsow

180

179.946706

0.120(1)

l82w

182

181.948205

26.498(29)

1 8 3 ~

183

182.950224

14.314(4)

1 8 4 ~

184

183.950932

30.642(8)

Is0Hf I

I '''Re

186 185

1

185.954362 184.952955

1

28.426(37) 37.40(2)

I

195

Appendix 2 Isotope

Nominal mass

Mass

"'Re

187

186.955751

62.60(2)

1840~

184

183.952491

0.020(3)

lS6OS

186

185.953838

1.58(10)

1870s

187

186.955748

1.6(1)

l=Os 1890~

1900~ 1920~

I

1

I 1 I

188 189 190 192 191

I

I

1 I I

Relative abundance

187.955836

I

13.3(2)

188.958145

I

16.1(3)

189.958445

26.4(4)

191.961479

41.0(3)

190.960591

37.3(5)

1931~

193

192.962923

62.7(5 )

Isopt

190

189.95993

0.01(1)

192Pt

I92

191.961035

0.79(6)

194Pt

194

193.962663

32.9(6)

195Pt

195

194.964774

33.8(6)

196Pt

196

195.964934

25.3(6)

I9*Pt

I

198

I

197.967875

I

7.2(2)

197Au

197

196.966551

-100

196Hg

196

195.965814

0.15(1 )

19'Hg

198

197.966752

9.97(8)

198.968262

16.87(10)

199.968309

23.10(16)

200.970285

13.18(8)

199Hg 200Hg 201Hg

1 1 I

199 200 201

I 1 I

202Hg

202

201.970625

29.86(20)

204Hg

204

203.973475

6.87(4)

203~1

203

202.972329

29.524( 14)

205~1

205

204.974412

70.476(14)

Appendix 2

196 Isotope

204Pb

1

I

Nominal mass 204

I

1

Mass 203.973028

I I

Relative abundance I .4(1)

206Pb

206

205.974449

24.1(1)

*"Pb

207

206.975880

22.1(1)

208Pb 209~i

232Th 2 3 4 u

235u

2 3 8 ~

I 1

1

I

1

208 209

I

I

207.976636 208.980384

I

I

52.4(1)

-100 ~-

232 234 235 238

100

1

I

234.040945 235.043922 238.050784

I

I

0.0055(5)

0.720(1) 99.2745( 15)

(Data source: Table of the Isotopes (Revised 1998), Norman E. Holden, Brookhaven National Laboratory, New York U.S.A. Data edited and compiled by Jason W.H. Wong, University of Sydney)

APPENDIX 3

Comparison of Common Ionisation Techniques* Ionisation t e c hique

Symbol

Mass limit

Types of molecules

Advantages

Electron ionisation

EI

500

Volatile organics, gases, non-polar compounds

Provides structural information as well as molecular weight

Chemical ionisation

CI

500

Volatile organics, gases, non-polar compounds

Easy to implement on EI source, enhances molecular ion production

Fast atom bombardment

FAB

20,000

Polar compounds including small biopolymers

Modest ionisation efficiencies for large molecular weight compounds

up to 1,000,000

Polar compounds from small molecules to large biopolymers

Easy to perform, suited to high throughput, high ionisation efficiencies

up to 5,000,000

Polar compounds from small molecules to large biopol ymers

High ionisation efficiencies, compatible with LC and CE separation

Matrix-assisted MALDI laser desorption ionisation

Electrospray ionisation

ESI

* Descriptions are to be used as a guide only.

197

APPENDIX 4

Comparison of the Performance of Mass Analysers* Analyser

Symbol

Dynamic mass range

Mass resolution

Advantages

Time-of-flight

TOF

unlimited

500 (linear) 1000 (reflector) 5,000 (with timelag focusing)

Relatively easy to construct, Inexpensive, non-scanning, high ion transmission

Magnetic sector B

10,000

10,000

High resolution, high mass accuracy

Quadrupole

Q

up to 5000

2000

Inexpensive. tolerant of higher pressures, fastscanning capability

Quadrupole ion trap

QIT

up to 5000

2000

Inexpensive, compact, fast-scanning capability

Ion cyclotron resonance

ICR

10,000

up to 500,000

Very high resolution and mass accuracy

*Values are to be used as a guide only. Values vary among instruments. Mass resolutions quoted represent 10% valley definitions

198

1,2, 15,29,43 etc. 18 28 as above, and 28,29 3 1,45, 59, etc. 17 15, 29, 43, 57, etc. 1, 15, 29, 43, 57, etc. 28,43, 57, etc.

Aliphatic alcohols

Phenols

Ethers

Aliphatic amines

Aldehydes and ketones

Carboxylic acids, esters and amides

19, 35/37, 79/81, 129 20, 36/38 15, 29, 43, 57, etc.

~~

etc.

-Fa,-Cl*, -Br*, -I' (favoured for Br and I) -HF, -HCl -CH,', -CH,CH,', -CH,(CH,),', etc. from cleavage a to the halide atom

-CH,=CH,, -R-CH2=CH2, etc. via a McLafferty rearrangement -OH (acid), - OR (ester) or -NH2 (amide)

-H',-CH,',-CH,CH,',-CH,(CH,),', (adjacent to C=O group)

-NH3 -CH3', -CH,CH2', -CH,(CH,),', etc.

-CH,O, -CH,CH,O', -CH3(CH2),0', etc.

-CO, -CHO'

-Ho, 4 3 2 , -CH3', -CH,CH,', etc. -H,O -CH,=CH,

-CH,', -CH,CH,',-CH,(CH,),',etc. -CHz=CHz

Formula

* Fragmentation processes to be used as a guide for the interpretation of EI mass spectra. Refer to Chapter 5 for more detail

Alkyl halides

15, 29,43, 57, etc. 28

Aliphatic hydrocarbons

~~

16 or 17 or 31,45, 59, etc.

Nominal mass loss

Compound class

Common Neutral Losses during the Fragmentation of Organic Compounds*

APPENDIX 5

APPENDIX 6

Summary of Common Fragment Ions Detected for Organic Compounds by Class* Compound class

Representative formula

mlz of fragment ions

Aliphatic hydrocarbons

CnH2n+2

15, 29,43, 57, 71, 85,99, 113

Aliphatic alcohols or ethers

C,H,,+,OH/R

31,45, 59, 73, 87, 101

Aliphatic amines

CnH2n+lNH2

30,44, 58,72, 86, 100, 114

Aldehydes and ketones

CnHZn+, COHIR

29,43, 57,71,85, 99

Carboxylic acids and esters

C,H,,-,O,H/R

45 (acid only), 59, 73, 87, 101

Amides

CnH2n+,CONH,

44, 58, 72, 86, 100

Alkyl halides

CnH2n+,X

33(F), 49/51(Cl), 93195(Br), 143(I) 42, 56, 70, 84,98, 112

* Fragment ions are to be used as a guide for the interpretation of EI mass spectra. Refer to Chapter 5 for more detail.

200

APPENDIX 7

Gas Phase Acidity Data Acid M_H

MeCgCH MeOH EtOH HC-CH iPrOH iBuOH tBuOH tBuCH,OH HF PhCH,OH nPrOH PhNH, CH,SO,CH, CH,CHO CF,CH,Og pyrrole MeSH EtSH nPrSH tBuSH H2S HCN

(kJ mol) 1589.31 1587.63 1574.66 1571.72 1566.28 1563.35 1562.93 1556.65 1555.40 1547.44 1541.16 1536.97 1534.88 1534.04 1525.67 1510.18 1503.06 1496.36 1 492.17 1485.06 1479.62 1478.36

(Data sourced from J.E. Bartmess and R.T. McIver Jr., The Gas Phase Acidity Scale, in Gus Phase Zon Chemistry, Vol. 2, Academic Press, New York, 1979. Data is derived from pulsed ICR measurements at 298 K and high pressure MS measurements (shaded) at 500 600 K)

20 1

APPENDIX 8

Amino Acid Residue Masses and Modifying Groups Residue

Code

Elemental compost ion

Monoisotopic muss

Average muss

Alanine Cysteine Aspartic acid Glutamic acid Phenylalanine Glycine Histidine Isoleucine Lysine Leucine Methionine Asparagine Proline Glutamine Arginine Serine Threonine Valine Tryptophan Tyrosine

A C D E F

C3H5N0 C3H5NOS C4H5N03 C5H7NO3 C,H9N0 C2H3N0

C5H9N0 CIIH,ON20 C9H9NO2

71.03711 103.00919 115.02694 129.04259 147.06841 57.02146 137.05891 113.08406 128.09496 1 13.08406 13 1.04049 114.04293 97.05276 128.05858 156.10111 87.03203 101.04768 99.06841 186.07931 163.06333

7 1.0788 103.1448 115.0886 129.1155 147.1766 57.0520 137.1412 113.1595 128.1742 113.1595 131.1986 114.1039 97.1 167 128.1742 156.1876 87.0782 101.1051 99.1326 186.2133 163.1760

Elemental composition

Monoisot op ic mass

A verage muss

1.00782 15.02347 29.00274 43.01839 57.07042 17.00274 16.01872 15.99491 79.96633 79.95682 57.02146

1.0079 15.0348 29.0183 43.0452 57.1 154 17.0073 16.0226 15.9994 79.9799 80.0642 57.0520

Modifving group

G H I K L M N P

Q

R

S T V

w Y

C6H7N30 C6H

11

No

C6H 12N20

I I No C,H9NOS C4H6N202 C,H,NO C,H*N202 C6H

C6H12N40

C3H5N02 C4H7N02

~~

Hydrogen Methyl Formyl Acetyl t-Butyl H ydroxyl Amide Oxidation Phosphorylation Sulphation Carboxyamidomethyl

202

APPENDIX 9

Mononucleotide Residue Masses Monois0top ic muss

Average mass

loH 12N506P

329.05252

329.2091

1OH I 2N50,P

345.04744

345.2085

C9H12N307P

305.04129

305.1841

U

C9H

306.02530

306.1688

Deoxyadenosine

dA

CIOHI3 N 2 W

3 13.05761

3 13.2097

Deoxyguanosine

dG

329.05252

329.2091

Deoxycytidine

dC

C9H12N306P

289.04637

289.1847

Thymidine

T

CIOHI&OP

304.04604

304.1963

Nucleotide

Code

Adenosine

A

Guanosine

G

Cytosine

C

Uracil

Elemental composition

,OH

1 lN203P

I 2N305P

203

APPENDIX 10

Monosaccharide Residue Masses Residue

Abbreviation

Monoisotopic muss

Average muss

Pentoses

Ara Rib XYl

132.04226

132.1161

Deox yhexoses

Fuc Rha

146.05791

146.1430

Hexosamines

GalN GlcN

161.06881

161.1577

Hexoses

Fru Gal Glc Man

162.05282

162.1424

Glucuronic acid

HexA

176.0321

176.1259

N -Acet ylhexosamines

CalNAc GlcNAc

203.07937

203.1950

N-Acetylneuraminic acid

NeuAc

291.09542

29 1.2579

N-Glycolylneuraminic acid

NeuGc

307.09033

307.2573

204

APPENDIX 11

Web Sites on Mass Spectrometry Resources i-mass.com h t tp: llww w. i-mass. com Ion Source http:llwww. ionsource. com Spectroscopy Now (incorporating Base Peak) http:llwww. spectroscopynow. com

Societies American Society for Mass Spectrometry http:llwwwasms.org Australian & New Zealand Society for Mass Spectrometry http:llwww. lutrobe.edu.aulanzsmsl British Mass Spectrometry Society http:llwww. bmss. org. uk International Mass Spectrometry Society http:llwww. imss.nl Mass Spectrometry Society of Japan http:llwww. rnssj.jp

Chemical and Biological Online Tools Is0 tope Pattern Calculator http:llwww. shcfiuc. ukl-chemlchemputerlisotopes. html FindMod tool http:llau. expasy.orgltoolslJindmod1 GlycoSuite at Proteome Systems https:lltmat.proteomesystems. comlglycosuitel Mascot Search - Matrix Science http:llwww. matrixscience. comlsearchform-select. html Peptide Search at EMBL h t tp:Ilw ww.mann. embl-heidelberg. del GroupPageslPageLinklpep t idesearchpage. h tml Protein Prospector at UCSF http:llprospectoK ucsf.edu/ PROWL at Rockefeller http:llprowl. rockefeller edul Web Elements Periodic Table http:llwww. webelements. corn1

Tutorials ASMS - What is Mass Spectrometry? http:llwww. usms.orglwhatismsl BMSS - What is Mass Spectrometry? h t tp:llw ww. bmss. org.uklwhut-islwha tisframese t.h tml Cambridge University tutorial http:llwww-methods. ch. cam. ac. uklmethlmsltheoryl i-mass guides http:llwww. i-mass. comlguidel

205

Subject Index ab initio molecular orbital calculations, 104 accelerator mass spectrometry, also AMS, 175 accurate mass measurement, 84 adiabatic transition, 14 alcohols, 93 aldehydes, 97 algorithms, 60 amides, 98 amines, 96 analog-to-digital converters, 60 antibody-peptide complex, MS detection of, 131 Apollo missions, 169 appearance energy, 15, 87 appearance potential, 24 array detectors, 57 Aston, Francis, 3,4 atmospheric pressure chemical ionisation, also APCI, 35 atom gun, 28 automatic gain control, 50 average mass, 4

bond resonance, 13 branching, of oligosaccharides, 144, 145 broadband excitation, 53 Budzikiewicz, Herbert, 90 capillary electrophoresis mass spectrometry, also CE-MS, 38 Caprioli, Richard, 134 carboxylic acids, 98 cathode ray tubes, 1,2 charge localisation, 86 charge remote, 86 charge reversal, or charge stripping, 71 charge transfer, also charge exchange, 26, 105 chemical ionisation, 25 chemical libraries, use in drug discovery and screening, 148, 155 chiral, kinetic method determination of enantiomeric composition, 153 cocaine, 150 collector, 55 collisional activation, 68 collisional activation theory, 69 collisionally activated dissociation, also CAD or CID, 69 comet, composition of, I70 COMPLX, algorithm, 131 computer acquisition, 59 continuous dynode, 56 continuous flow fast atom bombardment, also CF-FAB, 29 corpuscles, 1

Barber, Michael, 27, 114 base peak, 12 Beagle2, 170 P-lactam antibiotic, MS analysis of, 151 biomarkers, identification of, 148 biomolecular interaction analysis, also BIA, 156 bond dissociation energies, also BDE, 24,87, 105 206

Subject Index

cotinine, 102, 103 Coulomb, 10 cryogenic pump, or cryo, 64 cyclotron frequency, 5 1 data processing, 60 deconvoluted mass spectrum, 35 delayed extraction, also pulsed ion extraction, 42 delocalisation, 13 derivatisation, of oligosaccharides, 144 detector response, calibration of, 101 detectors, 55 diffusion pump, 62 direct current, DC, 47 discrete dynode, 56 dissociation, ion, 68 dissociative electron capture, 23 distonic ion, 89 Dole, Malcolm, 33 double bond equivalents, DBE, 7 double focusing, 45 dried droplet evaporation, 31 drug screening, by mass spectrometry, 156 Edman degradation, 118 electron, 1 electron affinity, 104 electron capture dissociation, also ECD, 73 electron impact, see electron ionisation, 10, 23 electron ionisation, EI, 10, 23 electron multiplier, also EM, 56 electrospray ionisation, also ESI, 33 elemental analysis, 157, 160 elemental composition, 6 elm, 3 environmental analysis, 160 epitope, identification by mass spectrometry, 131, 137 esters, 98 ethers, 95

207 even-electron rule, 87 excitation chirp, 53 exoglycosidase digestion, 144 Faraday cup, 55 fast atom (or ion) bombardment, also FAB, 27 Fenn, John, 33, 113 field desorption ionisation, also FD, 26 field-free region, 40 flowing afterglow, 107 forward geometry, 45 Fourier transform, also FT, 53 fragment ion linked scan, on magnetic sector mass analyser, 75 fragment ions, also daughter or product ions, 11,67 functional proteomics, 132 gas chromatography-mass spectrometry, also GC-MS, 26 gas phase acidity, 105, 106 genomics, 148 glycoconjugates, 140 Goldstein, Eugen, 1 halides, 99 hard ionisation, 25, 86 heterogeneity, in glyconjugates, 142 heterolytic bond cleavage, 88 high energy (keV) collision, 70 homolytic bond cleavage, 87 human hair, analysis of cocaine and, 6-acetylmorphine in, 150 hybrid instrument, 40, 54 hydrocarbons, 90 hydrogenldeuterium exchange MS, 125 image current, 52 imaging, mass spectrometric, 134, 136 inductive effect, 13

208

inductively coupled plasma mass spectrometry, also ICP-MS, 157, 161 influenza, mass spectrometric surveillance of, 131 inhibitor, MS analysis of complex with protein, 152 interpretation of MS/MS spectra, of peptides, 121 ion cyclotron resonance, also ICR, 51 ion microscopy, 171 ion mobility, of protein ions, 127 ion molecule reaction, rates of, 110 ion pair formation, 23 ion reflector, also reflectron or ion mirror, 41 ion source, 22 ion spray, see ESI, 33 ionisation energy, 14,105 ionisation potential, 15, 24 ion-molecule reactions, 108 ion-neutral complexes, 110 ionosphere, 167, 168 isotope dilution method, 162 isotope ratio mass spectrometry, also IRMS, 163 isotope-coded affinity tags, also ICAT, 135 isotopes, 4 ketones, 97 kinetic isotope effects, 111 kinetic method, 108 ladder sequencing, 1 18 laser desorption, also LD, 30 linked scans, 75 liquid chromatography mass spectrometry, also LC-MS, 36 low energy (keV) collision, 70

mlz, also mass-to-charge ratio, 11 magic bullet, 29 magnetic field strength, also B, 43

Subject Index

magnetic sector, also B, 43 MALDI matrices, 31 Mars, Viking and European Express missions, 168, 169 Mascot, 116 mass accuracy, 6, 19 mass analysers, 39 mass mapping, also mass fingerprinting, 1 15 mass resolution, 18 mass selective stability mode, 49 mass spectral databases, 86 mass spectrograph, 4 mass spectrometric immunoassay, also MSIA, 159 mass-to-charge ratio, also rnlz, 11 matrix-assisted laser desorption ionisation, MALDI, 30 McLafferty rearrangement, 89 McLafferty, Fred, 120 mean free path, 22 membrane introduction mass spectrometry, also MIMS, 166 mesomeric effect, 13 metabolic pathways, 154 metabolomics, 148 metastable ion, denoted m*, 16 microchannel plate, also MCP, 57 micro-electrospray, also nanospray, 39 microorganisms, direct MS analysis of, 135 MIKES, also mass analysed ion kinetic energy spectrum, 46, 73 modified nucleosides, 139 molecular ion, 10 molecular weight, 6 momentum, ion, 44 monoisotopic mass, 4 MSIMS, 67 multiple reaction monitoring, also MRM, 153 mustard gas, 166, 167 nanospray, 38

Subject Index neutral loss linked scan, on magnetic sector mass analyser, 76 neutralisation/reionisation, also NR, 71 nitrogen rule, 7 nominal molecular weight, 7 nucleic acids, 138 oligonucleotides, 138 oligonucleotides, sequencing of, 139 oligosaccharides, 140 oligosaccharides, sequencing of, 142 ortho-effect, 99 oxidative and reductive cleavage, of oligosaccharides, 144, 145, 146 Paul, Wolfgang, 47 peak matching, 84 peak of lightest isotopes, 19 peptide mass map, 115 peptide sequencing, and protein, 117 peptides, 114 pesticides, organic, 162 photodiode array, 58 photon-induced dissociation, also PID, 72 plasma desorption ionisation, also PD, 26 portable mass spectrometers, 165 positive rays, 1 post-translation modifications, 124 potential energy diagram, 109 precursor ion linked scan, on magnetic sector mass analyser, 76 PROFOUND, algorithm, 1 17 protein complexes and assemblies, 128 Protein Prospector, 116 protein structure and folding, 124 proteins, 114 proteomics, role of mass spectrometry in, 132, 148 proton affinity, 105, 106 PROWL, 116

209

QTOF, quadrupole time-of-flight tandem mass spectrometer, 82 quadrupole ion trap, or QIT or IT, 49 quadrupole, also Q or q, 47 qualitative analysis, of organic drugs and metabolites, 149 quantitation standards, 100 quantitation, of drugs and metabolites, 148, 153 quantitation, of organic compounds, 100 quasi-equilibrium theory, or QET, 14 quasi-molecular ion, also pseudomolecular ion, 11 radical cation, 10 radical-based studies, of protein structure, 128 radioactive tracers, 180 radiocarbon dating, 175, 177 radiofrequency, also RF, 47 reaction catalysts, analysis of, 172 reaction coordinate, 109 rearrangements, 89 reflecting time of flight, also rTOF, 41 reflectron, see reflector, 41 relative ion abundance, 4, 17 reverse geometry, 45 RF-only mode, quadrupole scan, 77 rotary pumps, 61 roughing pump, 61 secondary ion mass spectrometry, also SIMS, 27, 169, 171 selected ion monitoring, also SIM, 38, 101 selected reaction monitoring, also SRM, 153 SEQUEST, algorithm, 133 sigma(s)-bond cleavage, 88 soft ionisation, 25 space charge effects, 50

210 space, mass spectrometers in, 168 spray ionisation, 33 stability diagram, 48 stable isotope encoding, 155 Stoney, George, 1 streptavidin, MS detection of tetrameric complex of, 130 surface-enhanced laser desorption ionisation, also SELDI, 31 surface-induced dissociation, also SID, 72 SWISS-Prot database, 116 Tanaka, Koichi, 113 tandem accelerator, 175, 176 tandem mass spectrometric sequencing, of peptides, 119 tandem mass spectrometry, 67 Taylor cone, 33 thermal ionisation mass spectrometry, also TIMS, 160

Subject Index thermospray ionisation, also TSP, 33 Thomson, Joseph John, 1 time of flight, also TOF, 40 time-lag focusing, 42 TOFITOF, time of flight tandem mass spectrometer, 80 top down approach, to sequence proteins, 120 total ion current, 17 trace elements, nutrient analysis, 157 transition state, 109 triple quadrupole, 77 tropylium ion, 92 turbomolecular pump, or turbo, 63 Turin, shroud of, 178 vacuum pump, 61 velocity, ion, vertical transition, 13 Vestal, Marvin, 33 Wein, Wilhelm, 2